Model performance on MindsHub

See how models served through MindsHub Inference perform: response times, availability and cost, with current prices, API aliases and fallback routes.

Measurements refresh every 2 to 6 hours · About the data

All makers
Makers
Reasoning effort

Cost per task

Compare the cost of the same writing task. The model you choose as the reference is ×1; ×0.5 costs half as much. Each dot is an effort level. In Range per model, the ring marks the selected effort.

View

How to read the cost chart

Thick line: Range from the cheapest to the most expensive effort level.Dot: one effort level.Ring: the effort selected in the toolbar.Vertical line: the reference model’s cost, ×1. Dot: one effort level.Vertical line: the reference model’s cost, ×1.

Select a dot for estimated dollar cost, token counts, dates and tests.

Showing 46 models

Measurements: Oct 15 – Nov 18, 2026 UTC.

Measurements: Oct 15 – Nov 18, 2026 UTC.

×1 is Claude Opus 5.5 · max, the most expensive task on the board, the Leaderboard’s 100%; across the board the cheapest level costs ×0.00065. Every model on the same writing task, at the catalog’s list prices.

Each model’s range across effort levels. Ordered by cost at the selected effort.

All effort levels, grouped by model. Dollar estimates use current standard rates, without cache discounts.

  • MindsHub Aireffort: none to mediumMindsHub Air · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.00065Est. cost / task$0.0006Mean input tokens970Mean output tokens909Tests40IntervalOct 20 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.MindsHub Air · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.00078Est. cost / task$0.0007Mean input tokens970Mean output tokens1,126Tests40IntervalOct 20 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.MindsHub Air · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.00084Est. cost / task$0.0007Mean input tokens970Mean output tokens1,224Tests40IntervalOct 20 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00065–0.00084, none ×0.00065, low ×0.00078, medium ×0.00084
  • GPT-OSS 120B (Fireworks)no effort settingGPT-OSS 120B (Fireworks) · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.0016Est. cost / task$0.0014Mean input tokens1,032Mean output tokens2,012Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0016
  • GPT-6 Lunaeffort: none to maxGPT-6 Luna · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.00065Est. cost / task$0.0006Mean input tokens970Mean output tokens908Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6 Luna · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.00078Est. cost / task$0.0007Mean input tokens970Mean output tokens1,125Tests40IntervalOct 22 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6 Luna · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.00084Est. cost / task$0.0007Mean input tokens970Mean output tokens1,227Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6 Luna · effort: highRelative cost to Claude Opus 5.5 · max effort×0.00092Est. cost / task$0.0008Mean input tokens970Mean output tokens1,361Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6 Luna · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.00099Est. cost / task$0.0008Mean input tokens970Mean output tokens1,485Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6 Luna · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.0019Est. cost / task$0.0016Mean input tokens970Mean output tokens2,934Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00065–0.0019, none ×0.00065, low ×0.00078, medium ×0.00084, high ×0.00092, xhigh ×0.00099, max ×0.0019
  • GLM 5.3 Flasheffort: low to maxGLM 5.3 Flash · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.00077Est. cost / task$0.0007Mean input tokens995Mean output tokens1,008Tests40IntervalOct 22 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GLM 5.3 Flash · effort: highRelative cost to Claude Opus 5.5 · max effort×0.00083Est. cost / task$0.0007Mean input tokens994Mean output tokens1,108Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GLM 5.3 Flash · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.0049Est. cost / task$0.0042Mean input tokens995Mean output tokens8,009Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00077–0.0049, low ×0.00077, high ×0.00083, max ×0.0049
  • Muse Ember 1.0no effort settingMuse Ember 1.0 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.0055Est. cost / task$0.0046Mean input tokens986Mean output tokens9,079Tests40IntervalNov 10 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0055
  • Gemini 3.1 Flash-Liteeffort: minimal to highGemini 3.1 Flash-Lite · effort: minimalRelative cost to Claude Opus 5.5 · max effort×0.0025Est. cost / task$0.0021Mean input tokens999Mean output tokens1,265Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Gemini 3.1 Flash-Lite · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.003Est. cost / task$0.0026Mean input tokens999Mean output tokens1,542Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Gemini 3.1 Flash-Lite · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.0054Est. cost / task$0.0045Mean input tokens999Mean output tokens2,853Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Gemini 3.1 Flash-Lite · effort: highRelative cost to Claude Opus 5.5 · max effort×0.0074Est. cost / task$0.0062Mean input tokens999Mean output tokens3,983Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0025–0.0074, minimal ×0.0025, low ×0.003, medium ×0.0054, high ×0.0074
  • GPT 5.4 Nanoeffort: none to xhighGPT 5.4 Nano · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.0029Est. cost / task$0.0024Mean input tokens970Mean output tokens1,780Tests40IntervalOct 22 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT 5.4 Nano · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.0029Est. cost / task$0.0024Mean input tokens970Mean output tokens1,790Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT 5.4 Nano · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.0031Est. cost / task$0.0026Mean input tokens970Mean output tokens1,950Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT 5.4 Nano · effort: highRelative cost to Claude Opus 5.5 · max effort×0.0034Est. cost / task$0.0029Mean input tokens970Mean output tokens2,127Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT 5.4 Nano · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.0089Est. cost / task$0.0075Mean input tokens970Mean output tokens5,831Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0029–0.0089, none ×0.0029, low ×0.0029, medium ×0.0031, high ×0.0034, xhigh ×0.0089
  • DeepSeek V4.1 Flash 1017effort: none to maxDeepSeek V4.1 Flash 1017 · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.0018Est. cost / task$0.0015Mean input tokens980Mean output tokens1,040Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.DeepSeek V4.1 Flash 1017 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.0047Est. cost / task$0.0040Mean input tokens1,005Mean output tokens3,076Tests40IntervalOct 28 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.DeepSeek V4.1 Flash 1017 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.0056Est. cost / task$0.0047Mean input tokens1,005Mean output tokens3,698Tests40IntervalOct 27 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.DeepSeek V4.1 Flash 1017 · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.0097Est. cost / task$0.0082Mean input tokens1,005Mean output tokens6,559Tests40IntervalOct 28 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0018–0.0097, none ×0.0018, low ×0.0047, high ×0.0056, max ×0.0097
  • Claude Haiku 4.5no effort settingClaude Haiku 4.5 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.011Est. cost / task$0.0091Mean input tokens1,073Mean output tokens1,615Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.011
  • Gemini 3 Flash Previeweffort: high onlyGemini 3 Flash Preview · effort: highRelative cost to Claude Opus 5.5 · max effort×0.018Est. cost / task$0.0155Mean input tokens999Mean output tokens4,989Tests40IntervalOct 15 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.018, minimal not measured, low not measured, medium not measured, high ×0.018
  • Grok 4.5effort: high onlyGrok 4.5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.02Est. cost / task$0.0175Mean input tokens1,482Mean output tokens2,422Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.02, low not measured, medium not measured, high ×0.02
  • Claude Haiku 5.5effort: low to maxClaude Haiku 5.5 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.0012Est. cost / task$0.0011Mean input tokens1,489Mean output tokens1,806Tests40IntervalNov 13 – 14, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Haiku 5.5 · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.0014Est. cost / task$0.0012Mean input tokens1,489Mean output tokens2,051Tests40IntervalNov 7 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Haiku 5.5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.0019Est. cost / task$0.0016Mean input tokens1,489Mean output tokens2,905Tests40IntervalNov 13 – 14, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Haiku 5.5 · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.003Est. cost / task$0.0025Mean input tokens1,489Mean output tokens4,756Tests40IntervalNov 13 – 14, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Haiku 5.5 · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.024Est. cost / task$0.0204Mean input tokens1,489Mean output tokens40,462Tests40IntervalNov 13 – 14, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0012–0.024, low ×0.0012, medium ×0.0014, high ×0.0019, xhigh ×0.003, max ×0.024
  • Muse Spark 1.2no effort settingMuse Spark 1.2 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.03Est. cost / task$0.0253Mean input tokens986Mean output tokens5,657Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.03
  • Muse Spark 1.1no effort settingMuse Spark 1.1 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.033Est. cost / task$0.0281Mean input tokens986Mean output tokens6,311Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.033
  • MindsHub Blazeeffort: none to highMindsHub Blaze · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.0038Est. cost / task$0.0032Mean input tokens1,030Mean output tokens1,490Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.MindsHub Blaze · effort: highRelative cost to Claude Opus 5.5 · max effort×0.033Est. cost / task$0.0281Mean input tokens1,066Mean output tokens18,164Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0038–0.033, none ×0.0038, high ×0.033
  • Kimi K3.1 Prono effort settingKimi K3.1 Pro · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.036Est. cost / task$0.0308Mean input tokens1,056Mean output tokens2,873Tests40IntervalOct 27 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.036
  • Gemini 3.8 Flasheffort: low to highGemini 3.8 Flash · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.0094Est. cost / task$0.0080Mean input tokens999Mean output tokens1,925Tests40IntervalOct 22 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Gemini 3.8 Flash · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.026Est. cost / task$0.0217Mean input tokens999Mean output tokens5,584Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Gemini 3.8 Flash · effort: highRelative cost to Claude Opus 5.5 · max effort×0.043Est. cost / task$0.0362Mean input tokens999Mean output tokens9,448Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0094–0.043, low ×0.0094, medium ×0.026, high ×0.043
  • Muse Spark 1.3no effort settingMuse Spark 1.3 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.047Est. cost / task$0.0399Mean input tokens986Mean output tokens9,108Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.047
  • GLM 5.3effort: low to maxGLM 5.3 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.008Est. cost / task$0.0068Mean input tokens995Mean output tokens1,226Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GLM 5.3 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.008Est. cost / task$0.0068Mean input tokens994Mean output tokens1,229Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GLM 5.3 · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.052Est. cost / task$0.0443Mean input tokens995Mean output tokens9,755Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.008–0.052, low ×0.008, high ×0.008, max ×0.052
  • Kimi K3no effort settingKimi K3 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.053Est. cost / task$0.0450Mean input tokens992Mean output tokens2,803Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.053
  • GPT-6.1 Soleffort: low to maxGPT-6.1 Sol · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.017Est. cost / task$0.0143Mean input tokens970Mean output tokens1,234Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6.1 Sol · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.018Est. cost / task$0.0151Mean input tokens970Mean output tokens1,315Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6.1 Sol · effort: highRelative cost to Claude Opus 5.5 · max effort×0.021Est. cost / task$0.0180Mean input tokens970Mean output tokens1,610Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6.1 Sol · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.029Est. cost / task$0.0241Mean input tokens970Mean output tokens2,217Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6.1 Sol · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.056Est. cost / task$0.0476Mean input tokens970Mean output tokens4,569Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.017–0.056, low ×0.017, medium ×0.018, high ×0.021, xhigh ×0.029, max ×0.056
  • GPT 5.3 Codexeffort: low to xhighGPT 5.3 Codex · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.027Est. cost / task$0.0226Mean input tokens970Mean output tokens1,495Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT 5.3 Codex · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.029Est. cost / task$0.0245Mean input tokens970Mean output tokens1,632Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT 5.3 Codex · effort: highRelative cost to Claude Opus 5.5 · max effort×0.039Est. cost / task$0.0327Mean input tokens970Mean output tokens2,212Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT 5.3 Codex · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.063Est. cost / task$0.0530Mean input tokens970Mean output tokens3,663Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.027–0.063, low ×0.027, medium ×0.029, high ×0.039, xhigh ×0.063
  • GPT-6.1 Terraeffort: none to maxGPT-6.1 Terra · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.0071Est. cost / task$0.0060Mean input tokens970Mean output tokens1,014Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6.1 Terra · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.0071Est. cost / task$0.0060Mean input tokens970Mean output tokens998Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6.1 Terra · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.0078Est. cost / task$0.0066Mean input tokens970Mean output tokens1,120Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6.1 Terra · effort: highRelative cost to Claude Opus 5.5 · max effort×0.0097Est. cost / task$0.0082Mean input tokens970Mean output tokens1,439Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6.1 Terra · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.021Est. cost / task$0.0179Mean input tokens970Mean output tokens3,387Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6.1 Terra · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.088Est. cost / task$0.0744Mean input tokens970Mean output tokens14,692Tests40IntervalOct 29 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0071–0.088, none ×0.0071, low ×0.0071, medium ×0.0078, high ×0.0097, xhigh ×0.021, max ×0.088
  • GPT 5.4 Minieffort: none to xhighGPT 5.4 Mini · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.0079Est. cost / task$0.0066Mean input tokens970Mean output tokens1,315Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT 5.4 Mini · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.0079Est. cost / task$0.0067Mean input tokens970Mean output tokens1,322Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT 5.4 Mini · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.0087Est. cost / task$0.0074Mean input tokens970Mean output tokens1,473Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT 5.4 Mini · effort: highRelative cost to Claude Opus 5.5 · max effort×0.014Est. cost / task$0.0117Mean input tokens970Mean output tokens2,432Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT 5.4 Mini · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.099Est. cost / task$0.0835Mean input tokens970Mean output tokens18,394Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0079–0.099, none ×0.0079, low ×0.0079, medium ×0.0087, high ×0.014, xhigh ×0.099
  • Gemini 4 Argoneffort: low to highGemini 4 Argon · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.023Est. cost / task$0.0192Mean input tokens1,034Mean output tokens1,712Tests40IntervalOct 17 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Gemini 4 Argon · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.023Est. cost / task$0.0195Mean input tokens1,034Mean output tokens1,747Tests40IntervalOct 17 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Gemini 4 Argon · effort: highRelative cost to Claude Opus 5.5 · max effort×0.1Est. cost / task$0.0885Mean input tokens1,034Mean output tokens8,646Tests40IntervalOct 17 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.023–0.1, low ×0.023, medium ×0.023, high ×0.1
  • Qwen3.8-2.4T-A95Beffort: low to xhighQwen3.8-2.4T-A95B · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.012Est. cost / task$0.0100Mean input tokens1,054Mean output tokens1,315Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Qwen3.8-2.4T-A95B · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.016Est. cost / task$0.0133Mean input tokens1,028Mean output tokens1,880Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Qwen3.8-2.4T-A95B · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.12Est. cost / task$0.0974Mean input tokens1,066Mean output tokens15,872Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.012–0.12, low ×0.012, medium ×0.016, xhigh ×0.12
  • Gemini 3.1 Pro Previeweffort: low to highGemini 3.1 Pro Preview · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.07Est. cost / task$0.0595Mean input tokens999Mean output tokens4,789Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Gemini 3.1 Pro Preview · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.09Est. cost / task$0.0763Mean input tokens999Mean output tokens6,193Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Gemini 3.1 Pro Preview · effort: highRelative cost to Claude Opus 5.5 · max effort×0.12Est. cost / task$0.1016Mean input tokens999Mean output tokens8,300Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.07–0.12, low ×0.07, medium ×0.09, high ×0.12
  • Grok 4.7effort: low to xhighGrok 4.7 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.019Est. cost / task$0.0177Mean input tokens2,230Mean output tokens2,214Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Grok 4.7 · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.027Est. cost / task$0.0244Mean input tokens2,230Mean output tokens3,316Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Grok 4.7 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.061Est. cost / task$0.0532Mean input tokens2,230Mean output tokens8,126Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Grok 4.7 · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.13Est. cost / task$0.1108Mean input tokens2,231Mean output tokens17,715Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.019–0.13, low ×0.019, medium ×0.027, high ×0.061, xhigh ×0.13
  • Muse Wonder 1.0no effort settingMuse Wonder 1.0 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.28Est. cost / task$0.2408Mean input tokens986Mean output tokens11,845Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.28
  • GPT-6 Astraeffort: low to maxGPT-6 Astra · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.087Est. cost / task$0.0735Mean input tokens970Mean output tokens1,276Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6 Astra · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.091Est. cost / task$0.0772Mean input tokens970Mean output tokens1,350Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6 Astra · effort: highRelative cost to Claude Opus 5.5 · max effort×0.11Est. cost / task$0.0951Mean input tokens970Mean output tokens1,709Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6 Astra · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.15Est. cost / task$0.1298Mean input tokens970Mean output tokens2,402Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.GPT-6 Astra · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.31Est. cost / task$0.2609Mean input tokens970Mean output tokens5,023Tests40IntervalNov 10 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.087–0.31, low ×0.087, medium ×0.091, high ×0.11, xhigh ×0.15, max ×0.31
  • Claude Sonnet 5.5effort: low to maxClaude Sonnet 5.5 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.021Est. cost / task$0.0177Mean input tokens1,489Mean output tokens1,475Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Sonnet 5.5 · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.021Est. cost / task$0.0180Mean input tokens1,489Mean output tokens1,503Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Sonnet 5.5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.025Est. cost / task$0.0215Mean input tokens1,489Mean output tokens1,852Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Sonnet 5.5 · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.033Est. cost / task$0.0281Mean input tokens1,489Mean output tokens2,516Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Sonnet 5.5 · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.38Est. cost / task$0.3207Mean input tokens1,489Mean output tokens31,777Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.021–0.38, low ×0.021, medium ×0.021, high ×0.025, xhigh ×0.033, max ×0.38
  • Claude Fable 5.1effort: low to maxClaude Fable 5.1 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.12Est. cost / task$0.1000Mean input tokens1,489Mean output tokens1,702Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Fable 5.1 · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.13Est. cost / task$0.1089Mean input tokens1,489Mean output tokens1,880Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Fable 5.1 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.14Est. cost / task$0.1217Mean input tokens1,489Mean output tokens2,136Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Fable 5.1 · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.26Est. cost / task$0.2190Mean input tokens1,489Mean output tokens4,083Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Fable 5.1 · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.7Est. cost / task$0.5883Mean input tokens1,489Mean output tokens11,468Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.12–0.7, low ×0.12, medium ×0.13, high ×0.14, xhigh ×0.26, max ×0.7
  • Claude Opus 5.5effort: low to max · ×1 at maxClaude Opus 5.5 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.051Est. cost / task$0.0432Mean input tokens1,489Mean output tokens1,860Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Opus 5.5 · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.056Est. cost / task$0.0475Mean input tokens1,489Mean output tokens2,078Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Opus 5.5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.067Est. cost / task$0.0567Mean input tokens1,489Mean output tokens2,538Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Opus 5.5 · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.15Est. cost / task$0.1253Mean input tokens1,489Mean output tokens5,966Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.Claude Opus 5.5 · effort: maxRelative cost to Claude Opus 5.5 · max effort×1.0Est. cost / task$0.8452Mean input tokens1,489Mean output tokens41,962Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.051–1.0, low ×0.051, medium ×0.056, high ×0.067, xhigh ×0.15, max ×1.0
  • GPT 5.6 Lunaeffort: medium onlyGPT 5.6 Luna · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.0021Est. cost / task$0.0018Mean input tokens970Mean output tokens1,303Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0021, none not measured, low not measured, medium ×0.0021, high not measured, xhigh not measured, max not measured
  • DeepSeek V4.1 Flasheffort: high onlyDeepSeek V4.1 Flash · effort: highRelative cost to Claude Opus 5.5 · max effort×0.011Est. cost / task$0.0096Mean input tokens1,005Mean output tokens7,726Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.011, none not measured, low not measured, high ×0.011, max not measured
  • GPT-6 Soleffort: medium onlyGPT-6 Sol · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.018Est. cost / task$0.0153Mean input tokens970Mean output tokens1,333Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.018, none not measured, low not measured, medium ×0.018, high not measured, xhigh not measured, max not measured
  • GPT 5.6 Terraeffort: medium onlyGPT 5.6 Terra · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.02Est. cost / task$0.0173Mean input tokens970Mean output tokens1,278Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.02, none not measured, low not measured, medium ×0.02, high not measured, xhigh not measured, max not measured
  • Claude Sonnet 5effort: high onlyClaude Sonnet 5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.024Est. cost / task$0.0203Mean input tokens1,487Mean output tokens1,736Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.024, low not measured, medium not measured, high ×0.024, max not measured
  • Gemini 3.7 Flasheffort: medium onlyGemini 3.7 Flash · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.026Est. cost / task$0.0221Mean input tokens999Mean output tokens5,702Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.026, low not measured, medium ×0.026, high not measured
  • Gemini 3.6 Flasheffort: medium onlyGemini 3.6 Flash · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.029Est. cost / task$0.0244Mean input tokens999Mean output tokens6,303Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.029, minimal not measured, low not measured, medium ×0.029, high not measured
  • GPT 5.6 Soleffort: medium onlyGPT 5.6 Sol · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.053Est. cost / task$0.0447Mean input tokens970Mean output tokens1,328Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.053, none not measured, low not measured, medium ×0.053, high not measured, xhigh not measured, max not measured
  • Grok 4.6effort: high onlyGrok 4.6 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.059Est. cost / task$0.0504Mean input tokens1,624Mean output tokens7,865Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.059, low not measured, medium not measured, high ×0.059, xhigh not measured
  • Claude Opus 5effort: high onlyClaude Opus 5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.064Est. cost / task$0.0538Mean input tokens1,487Mean output tokens1,856Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.064, low not measured, medium not measured, high ×0.064, xhigh not measured, max not measured
  • Gemini 3.5 Flasheffort: medium onlyGemini 3.5 Flash · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.07Est. cost / task$0.0596Mean input tokens999Mean output tokens6,453Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.07, minimal not measured, low not measured, medium ×0.07, high not measured
  • Claude Fable 5.2effort: high onlyClaude Fable 5.2 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.1Est. cost / task$0.0864Mean input tokens1,489Mean output tokens1,863Tests40IntervalNov 10 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.1, low not measured, medium not measured, high ×0.1, xhigh not measured, max not measured
  • Claude Fable 5effort: high onlyClaude Fable 5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.12Est. cost / task$0.1027Mean input tokens1,487Mean output tokens1,756Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.12, low not measured, medium not measured, high ×0.12, xhigh not measured, max not measured
Model / effortRelative costEst. cost / task
    • GPT-6 Lunaeffort: noneGPT-6 Luna · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.00065Est. cost / task$0.0006Mean input tokens970Mean output tokens908Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00065Est. cost / task$0.0006
    • GPT-6 Luna, effort: lowGPT-6 Luna · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.00078Est. cost / task$0.0007Mean input tokens970Mean output tokens1,125Tests40IntervalOct 22 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00078Est. cost / task$0.0007
    • GPT-6 Luna, effort: mediumGPT-6 Luna · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.00084Est. cost / task$0.0007Mean input tokens970Mean output tokens1,227Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00084Est. cost / task$0.0007
    • GPT-6 Luna, effort: highGPT-6 Luna · effort: highRelative cost to Claude Opus 5.5 · max effort×0.00092Est. cost / task$0.0008Mean input tokens970Mean output tokens1,361Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00092Est. cost / task$0.0008
    • GPT-6 Luna, effort: xhighGPT-6 Luna · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.00099Est. cost / task$0.0008Mean input tokens970Mean output tokens1,485Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00099Est. cost / task$0.0008
    • GPT-6 Luna, effort: maxGPT-6 Luna · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.0019Est. cost / task$0.0016Mean input tokens970Mean output tokens2,934Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0019Est. cost / task$0.0016
    • MindsHub Aireffort: noneMindsHub Air · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.00065Est. cost / task$0.0006Mean input tokens970Mean output tokens909Tests40IntervalOct 20 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00065Est. cost / task$0.0006
    • MindsHub Air, effort: lowMindsHub Air · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.00078Est. cost / task$0.0007Mean input tokens970Mean output tokens1,126Tests40IntervalOct 20 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00078Est. cost / task$0.0007
    • MindsHub Air, effort: mediumMindsHub Air · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.00084Est. cost / task$0.0007Mean input tokens970Mean output tokens1,224Tests40IntervalOct 20 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00084Est. cost / task$0.0007
    • GLM 5.3 Flasheffort: lowGLM 5.3 Flash · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.00077Est. cost / task$0.0007Mean input tokens995Mean output tokens1,008Tests40IntervalOct 22 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00077Est. cost / task$0.0007
    • GLM 5.3 Flash, effort: highGLM 5.3 Flash · effort: highRelative cost to Claude Opus 5.5 · max effort×0.00083Est. cost / task$0.0007Mean input tokens994Mean output tokens1,108Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.00083Est. cost / task$0.0007
    • GLM 5.3 Flash, effort: maxGLM 5.3 Flash · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.0049Est. cost / task$0.0042Mean input tokens995Mean output tokens8,009Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0049Est. cost / task$0.0042
    • Claude Haiku 5.5effort: lowClaude Haiku 5.5 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.0012Est. cost / task$0.0011Mean input tokens1,489Mean output tokens1,806Tests40IntervalNov 13 – 14, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0012Est. cost / task$0.0011
    • Claude Haiku 5.5, effort: mediumClaude Haiku 5.5 · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.0014Est. cost / task$0.0012Mean input tokens1,489Mean output tokens2,051Tests40IntervalNov 7 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0014Est. cost / task$0.0012
    • Claude Haiku 5.5, effort: highClaude Haiku 5.5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.0019Est. cost / task$0.0016Mean input tokens1,489Mean output tokens2,905Tests40IntervalNov 13 – 14, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0019Est. cost / task$0.0016
    • Claude Haiku 5.5, effort: xhighClaude Haiku 5.5 · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.003Est. cost / task$0.0025Mean input tokens1,489Mean output tokens4,756Tests40IntervalNov 13 – 14, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.003Est. cost / task$0.0025
    • Claude Haiku 5.5, effort: maxClaude Haiku 5.5 · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.024Est. cost / task$0.0204Mean input tokens1,489Mean output tokens40,462Tests40IntervalNov 13 – 14, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.024Est. cost / task$0.0204
    • GPT-OSS 120B (Fireworks)no effort settingGPT-OSS 120B (Fireworks) · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.0016Est. cost / task$0.0014Mean input tokens1,032Mean output tokens2,012Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0016Est. cost / task$0.0014
    • DeepSeek V4.1 Flash 1017effort: noneDeepSeek V4.1 Flash 1017 · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.0018Est. cost / task$0.0015Mean input tokens980Mean output tokens1,040Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0018Est. cost / task$0.0015
    • DeepSeek V4.1 Flash 1017, effort: lowDeepSeek V4.1 Flash 1017 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.0047Est. cost / task$0.0040Mean input tokens1,005Mean output tokens3,076Tests40IntervalOct 28 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0047Est. cost / task$0.0040
    • DeepSeek V4.1 Flash 1017, effort: highDeepSeek V4.1 Flash 1017 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.0056Est. cost / task$0.0047Mean input tokens1,005Mean output tokens3,698Tests40IntervalOct 27 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0056Est. cost / task$0.0047
    • DeepSeek V4.1 Flash 1017, effort: maxDeepSeek V4.1 Flash 1017 · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.0097Est. cost / task$0.0082Mean input tokens1,005Mean output tokens6,559Tests40IntervalOct 28 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0097Est. cost / task$0.0082
    • GPT 5.6 Lunaeffort: none— not measured—Est. cost / task—
    • GPT 5.6 Luna, effort: low— not measured—Est. cost / task—
    • GPT 5.6 Luna, effort: mediumGPT 5.6 Luna · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.0021Est. cost / task$0.0018Mean input tokens970Mean output tokens1,303Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0021Est. cost / task$0.0018
    • GPT 5.6 Luna, effort: high— not measured—Est. cost / task—
    • GPT 5.6 Luna, effort: xhigh— not measured—Est. cost / task—
    • GPT 5.6 Luna, effort: max— not measured—Est. cost / task—
    • Gemini 3.1 Flash-Liteeffort: minimalGemini 3.1 Flash-Lite · effort: minimalRelative cost to Claude Opus 5.5 · max effort×0.0025Est. cost / task$0.0021Mean input tokens999Mean output tokens1,265Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0025Est. cost / task$0.0021
    • Gemini 3.1 Flash-Lite, effort: lowGemini 3.1 Flash-Lite · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.003Est. cost / task$0.0026Mean input tokens999Mean output tokens1,542Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.003Est. cost / task$0.0026
    • Gemini 3.1 Flash-Lite, effort: mediumGemini 3.1 Flash-Lite · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.0054Est. cost / task$0.0045Mean input tokens999Mean output tokens2,853Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0054Est. cost / task$0.0045
    • Gemini 3.1 Flash-Lite, effort: highGemini 3.1 Flash-Lite · effort: highRelative cost to Claude Opus 5.5 · max effort×0.0074Est. cost / task$0.0062Mean input tokens999Mean output tokens3,983Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0074Est. cost / task$0.0062
    • GPT 5.4 Nanoeffort: noneGPT 5.4 Nano · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.0029Est. cost / task$0.0024Mean input tokens970Mean output tokens1,780Tests40IntervalOct 22 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0029Est. cost / task$0.0024
    • GPT 5.4 Nano, effort: lowGPT 5.4 Nano · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.0029Est. cost / task$0.0024Mean input tokens970Mean output tokens1,790Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0029Est. cost / task$0.0024
    • GPT 5.4 Nano, effort: mediumGPT 5.4 Nano · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.0031Est. cost / task$0.0026Mean input tokens970Mean output tokens1,950Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0031Est. cost / task$0.0026
    • GPT 5.4 Nano, effort: highGPT 5.4 Nano · effort: highRelative cost to Claude Opus 5.5 · max effort×0.0034Est. cost / task$0.0029Mean input tokens970Mean output tokens2,127Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0034Est. cost / task$0.0029
    • GPT 5.4 Nano, effort: xhighGPT 5.4 Nano · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.0089Est. cost / task$0.0075Mean input tokens970Mean output tokens5,831Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0089Est. cost / task$0.0075
    • MindsHub Blazeeffort: noneMindsHub Blaze · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.0038Est. cost / task$0.0032Mean input tokens1,030Mean output tokens1,490Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0038Est. cost / task$0.0032
    • MindsHub Blaze, effort: highMindsHub Blaze · effort: highRelative cost to Claude Opus 5.5 · max effort×0.033Est. cost / task$0.0281Mean input tokens1,066Mean output tokens18,164Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.033Est. cost / task$0.0281
    • Muse Ember 1.0no effort settingMuse Ember 1.0 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.0055Est. cost / task$0.0046Mean input tokens986Mean output tokens9,079Tests40IntervalNov 10 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0055Est. cost / task$0.0046
    • GPT-6.1 Terraeffort: noneGPT-6.1 Terra · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.0071Est. cost / task$0.0060Mean input tokens970Mean output tokens1,014Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0071Est. cost / task$0.0060
    • GPT-6.1 Terra, effort: lowGPT-6.1 Terra · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.0071Est. cost / task$0.0060Mean input tokens970Mean output tokens998Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0071Est. cost / task$0.0060
    • GPT-6.1 Terra, effort: mediumGPT-6.1 Terra · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.0078Est. cost / task$0.0066Mean input tokens970Mean output tokens1,120Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0078Est. cost / task$0.0066
    • GPT-6.1 Terra, effort: highGPT-6.1 Terra · effort: highRelative cost to Claude Opus 5.5 · max effort×0.0097Est. cost / task$0.0082Mean input tokens970Mean output tokens1,439Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0097Est. cost / task$0.0082
    • GPT-6.1 Terra, effort: xhighGPT-6.1 Terra · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.021Est. cost / task$0.0179Mean input tokens970Mean output tokens3,387Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.021Est. cost / task$0.0179
    • GPT-6.1 Terra, effort: maxGPT-6.1 Terra · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.088Est. cost / task$0.0744Mean input tokens970Mean output tokens14,692Tests40IntervalOct 29 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.088Est. cost / task$0.0744
    • GPT 5.4 Minieffort: noneGPT 5.4 Mini · effort: noneRelative cost to Claude Opus 5.5 · max effort×0.0079Est. cost / task$0.0066Mean input tokens970Mean output tokens1,315Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0079Est. cost / task$0.0066
    • GPT 5.4 Mini, effort: lowGPT 5.4 Mini · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.0079Est. cost / task$0.0067Mean input tokens970Mean output tokens1,322Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0079Est. cost / task$0.0067
    • GPT 5.4 Mini, effort: mediumGPT 5.4 Mini · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.0087Est. cost / task$0.0074Mean input tokens970Mean output tokens1,473Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0087Est. cost / task$0.0074
    • GPT 5.4 Mini, effort: highGPT 5.4 Mini · effort: highRelative cost to Claude Opus 5.5 · max effort×0.014Est. cost / task$0.0117Mean input tokens970Mean output tokens2,432Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.014Est. cost / task$0.0117
    • GPT 5.4 Mini, effort: xhighGPT 5.4 Mini · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.099Est. cost / task$0.0835Mean input tokens970Mean output tokens18,394Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.099Est. cost / task$0.0835
    • GLM 5.3effort: lowGLM 5.3 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.008Est. cost / task$0.0068Mean input tokens995Mean output tokens1,226Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.008Est. cost / task$0.0068
    • GLM 5.3, effort: highGLM 5.3 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.008Est. cost / task$0.0068Mean input tokens994Mean output tokens1,229Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.008Est. cost / task$0.0068
    • GLM 5.3, effort: maxGLM 5.3 · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.052Est. cost / task$0.0443Mean input tokens995Mean output tokens9,755Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.052Est. cost / task$0.0443
    • Gemini 3.8 Flasheffort: lowGemini 3.8 Flash · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.0094Est. cost / task$0.0080Mean input tokens999Mean output tokens1,925Tests40IntervalOct 22 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.0094Est. cost / task$0.0080
    • Gemini 3.8 Flash, effort: mediumGemini 3.8 Flash · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.026Est. cost / task$0.0217Mean input tokens999Mean output tokens5,584Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.026Est. cost / task$0.0217
    • Gemini 3.8 Flash, effort: highGemini 3.8 Flash · effort: highRelative cost to Claude Opus 5.5 · max effort×0.043Est. cost / task$0.0362Mean input tokens999Mean output tokens9,448Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.043Est. cost / task$0.0362
    • Claude Haiku 4.5no effort settingClaude Haiku 4.5 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.011Est. cost / task$0.0091Mean input tokens1,073Mean output tokens1,615Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.011Est. cost / task$0.0091
    • DeepSeek V4.1 Flasheffort: none— not measured—Est. cost / task—
    • DeepSeek V4.1 Flash, effort: low— not measured—Est. cost / task—
    • DeepSeek V4.1 Flash, effort: highDeepSeek V4.1 Flash · effort: highRelative cost to Claude Opus 5.5 · max effort×0.011Est. cost / task$0.0096Mean input tokens1,005Mean output tokens7,726Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.011Est. cost / task$0.0096
    • DeepSeek V4.1 Flash, effort: max— not measured—Est. cost / task—
    • Qwen3.8-2.4T-A95Beffort: lowQwen3.8-2.4T-A95B · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.012Est. cost / task$0.0100Mean input tokens1,054Mean output tokens1,315Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.012Est. cost / task$0.0100
    • Qwen3.8-2.4T-A95B, effort: mediumQwen3.8-2.4T-A95B · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.016Est. cost / task$0.0133Mean input tokens1,028Mean output tokens1,880Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.016Est. cost / task$0.0133
    • Qwen3.8-2.4T-A95B, effort: xhighQwen3.8-2.4T-A95B · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.12Est. cost / task$0.0974Mean input tokens1,066Mean output tokens15,872Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.12Est. cost / task$0.0974
    • GPT-6.1 Soleffort: lowGPT-6.1 Sol · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.017Est. cost / task$0.0143Mean input tokens970Mean output tokens1,234Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.017Est. cost / task$0.0143
    • GPT-6.1 Sol, effort: mediumGPT-6.1 Sol · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.018Est. cost / task$0.0151Mean input tokens970Mean output tokens1,315Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.018Est. cost / task$0.0151
    • GPT-6.1 Sol, effort: highGPT-6.1 Sol · effort: highRelative cost to Claude Opus 5.5 · max effort×0.021Est. cost / task$0.0180Mean input tokens970Mean output tokens1,610Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.021Est. cost / task$0.0180
    • GPT-6.1 Sol, effort: xhighGPT-6.1 Sol · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.029Est. cost / task$0.0241Mean input tokens970Mean output tokens2,217Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.029Est. cost / task$0.0241
    • GPT-6.1 Sol, effort: maxGPT-6.1 Sol · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.056Est. cost / task$0.0476Mean input tokens970Mean output tokens4,569Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.056Est. cost / task$0.0476
    • GPT-6 Soleffort: none— not measured—Est. cost / task—
    • GPT-6 Sol, effort: low— not measured—Est. cost / task—
    • GPT-6 Sol, effort: mediumGPT-6 Sol · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.018Est. cost / task$0.0153Mean input tokens970Mean output tokens1,333Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.018Est. cost / task$0.0153
    • GPT-6 Sol, effort: high— not measured—Est. cost / task—
    • GPT-6 Sol, effort: xhigh— not measured—Est. cost / task—
    • GPT-6 Sol, effort: max— not measured—Est. cost / task—
    • Gemini 3 Flash Previeweffort: minimal— not measured—Est. cost / task—
    • Gemini 3 Flash Preview, effort: low— not measured—Est. cost / task—
    • Gemini 3 Flash Preview, effort: medium— not measured—Est. cost / task—
    • Gemini 3 Flash Preview, effort: highGemini 3 Flash Preview · effort: highRelative cost to Claude Opus 5.5 · max effort×0.018Est. cost / task$0.0155Mean input tokens999Mean output tokens4,989Tests40IntervalOct 15 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.018Est. cost / task$0.0155
    • Grok 4.7effort: lowGrok 4.7 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.019Est. cost / task$0.0177Mean input tokens2,230Mean output tokens2,214Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.019Est. cost / task$0.0177
    • Grok 4.7, effort: mediumGrok 4.7 · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.027Est. cost / task$0.0244Mean input tokens2,230Mean output tokens3,316Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.027Est. cost / task$0.0244
    • Grok 4.7, effort: highGrok 4.7 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.061Est. cost / task$0.0532Mean input tokens2,230Mean output tokens8,126Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.061Est. cost / task$0.0532
    • Grok 4.7, effort: xhighGrok 4.7 · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.13Est. cost / task$0.1108Mean input tokens2,231Mean output tokens17,715Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.13Est. cost / task$0.1108
    • Grok 4.5effort: low— not measured—Est. cost / task—
    • Grok 4.5, effort: medium— not measured—Est. cost / task—
    • Grok 4.5, effort: highGrok 4.5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.02Est. cost / task$0.0175Mean input tokens1,482Mean output tokens2,422Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.02Est. cost / task$0.0175
    • GPT 5.6 Terraeffort: none— not measured—Est. cost / task—
    • GPT 5.6 Terra, effort: low— not measured—Est. cost / task—
    • GPT 5.6 Terra, effort: mediumGPT 5.6 Terra · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.02Est. cost / task$0.0173Mean input tokens970Mean output tokens1,278Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.02Est. cost / task$0.0173
    • GPT 5.6 Terra, effort: high— not measured—Est. cost / task—
    • GPT 5.6 Terra, effort: xhigh— not measured—Est. cost / task—
    • GPT 5.6 Terra, effort: max— not measured—Est. cost / task—
    • Claude Sonnet 5.5effort: lowClaude Sonnet 5.5 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.021Est. cost / task$0.0177Mean input tokens1,489Mean output tokens1,475Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.021Est. cost / task$0.0177
    • Claude Sonnet 5.5, effort: mediumClaude Sonnet 5.5 · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.021Est. cost / task$0.0180Mean input tokens1,489Mean output tokens1,503Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.021Est. cost / task$0.0180
    • Claude Sonnet 5.5, effort: highClaude Sonnet 5.5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.025Est. cost / task$0.0215Mean input tokens1,489Mean output tokens1,852Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.025Est. cost / task$0.0215
    • Claude Sonnet 5.5, effort: xhighClaude Sonnet 5.5 · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.033Est. cost / task$0.0281Mean input tokens1,489Mean output tokens2,516Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.033Est. cost / task$0.0281
    • Claude Sonnet 5.5, effort: maxClaude Sonnet 5.5 · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.38Est. cost / task$0.3207Mean input tokens1,489Mean output tokens31,777Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.38Est. cost / task$0.3207
    • Gemini 4 Argoneffort: lowGemini 4 Argon · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.023Est. cost / task$0.0192Mean input tokens1,034Mean output tokens1,712Tests40IntervalOct 17 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.023Est. cost / task$0.0192
    • Gemini 4 Argon, effort: mediumGemini 4 Argon · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.023Est. cost / task$0.0195Mean input tokens1,034Mean output tokens1,747Tests40IntervalOct 17 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.023Est. cost / task$0.0195
    • Gemini 4 Argon, effort: highGemini 4 Argon · effort: highRelative cost to Claude Opus 5.5 · max effort×0.1Est. cost / task$0.0885Mean input tokens1,034Mean output tokens8,646Tests40IntervalOct 17 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.1Est. cost / task$0.0885
    • Claude Sonnet 5effort: low— not measured—Est. cost / task—
    • Claude Sonnet 5, effort: medium— not measured—Est. cost / task—
    • Claude Sonnet 5, effort: highClaude Sonnet 5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.024Est. cost / task$0.0203Mean input tokens1,487Mean output tokens1,736Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.024Est. cost / task$0.0203
    • Claude Sonnet 5, effort: max— not measured—Est. cost / task—
    • Gemini 3.7 Flasheffort: low— not measured—Est. cost / task—
    • Gemini 3.7 Flash, effort: mediumGemini 3.7 Flash · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.026Est. cost / task$0.0221Mean input tokens999Mean output tokens5,702Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.026Est. cost / task$0.0221
    • Gemini 3.7 Flash, effort: high— not measured—Est. cost / task—
    • GPT 5.3 Codexeffort: lowGPT 5.3 Codex · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.027Est. cost / task$0.0226Mean input tokens970Mean output tokens1,495Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.027Est. cost / task$0.0226
    • GPT 5.3 Codex, effort: mediumGPT 5.3 Codex · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.029Est. cost / task$0.0245Mean input tokens970Mean output tokens1,632Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.029Est. cost / task$0.0245
    • GPT 5.3 Codex, effort: highGPT 5.3 Codex · effort: highRelative cost to Claude Opus 5.5 · max effort×0.039Est. cost / task$0.0327Mean input tokens970Mean output tokens2,212Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.039Est. cost / task$0.0327
    • GPT 5.3 Codex, effort: xhighGPT 5.3 Codex · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.063Est. cost / task$0.0530Mean input tokens970Mean output tokens3,663Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.063Est. cost / task$0.0530
    • Gemini 3.6 Flasheffort: minimal— not measured—Est. cost / task—
    • Gemini 3.6 Flash, effort: low— not measured—Est. cost / task—
    • Gemini 3.6 Flash, effort: mediumGemini 3.6 Flash · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.029Est. cost / task$0.0244Mean input tokens999Mean output tokens6,303Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.029Est. cost / task$0.0244
    • Gemini 3.6 Flash, effort: high— not measured—Est. cost / task—
    • Muse Spark 1.2no effort settingMuse Spark 1.2 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.03Est. cost / task$0.0253Mean input tokens986Mean output tokens5,657Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.03Est. cost / task$0.0253
    • Muse Spark 1.1no effort settingMuse Spark 1.1 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.033Est. cost / task$0.0281Mean input tokens986Mean output tokens6,311Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.033Est. cost / task$0.0281
    • Kimi K3.1 Prono effort settingKimi K3.1 Pro · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.036Est. cost / task$0.0308Mean input tokens1,056Mean output tokens2,873Tests40IntervalOct 27 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.036Est. cost / task$0.0308
    • Muse Spark 1.3no effort settingMuse Spark 1.3 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.047Est. cost / task$0.0399Mean input tokens986Mean output tokens9,108Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.047Est. cost / task$0.0399
    • Claude Opus 5.5effort: lowClaude Opus 5.5 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.051Est. cost / task$0.0432Mean input tokens1,489Mean output tokens1,860Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.051Est. cost / task$0.0432
    • Claude Opus 5.5, effort: mediumClaude Opus 5.5 · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.056Est. cost / task$0.0475Mean input tokens1,489Mean output tokens2,078Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.056Est. cost / task$0.0475
    • Claude Opus 5.5, effort: highClaude Opus 5.5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.067Est. cost / task$0.0567Mean input tokens1,489Mean output tokens2,538Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.067Est. cost / task$0.0567
    • Claude Opus 5.5, effort: xhighClaude Opus 5.5 · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.15Est. cost / task$0.1253Mean input tokens1,489Mean output tokens5,966Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.15Est. cost / task$0.1253
    • Claude Opus 5.5, effort: max · ×1Claude Opus 5.5 · effort: maxRelative cost to Claude Opus 5.5 · max effort×1.0Est. cost / task$0.8452Mean input tokens1,489Mean output tokens41,962Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×1.0Est. cost / task$0.8452
    • GPT 5.6 Soleffort: none— not measured—Est. cost / task—
    • GPT 5.6 Sol, effort: low— not measured—Est. cost / task—
    • GPT 5.6 Sol, effort: mediumGPT 5.6 Sol · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.053Est. cost / task$0.0447Mean input tokens970Mean output tokens1,328Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.053Est. cost / task$0.0447
    • GPT 5.6 Sol, effort: high— not measured—Est. cost / task—
    • GPT 5.6 Sol, effort: xhigh— not measured—Est. cost / task—
    • GPT 5.6 Sol, effort: max— not measured—Est. cost / task—
    • Kimi K3no effort settingKimi K3 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.053Est. cost / task$0.0450Mean input tokens992Mean output tokens2,803Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.053Est. cost / task$0.0450
    • Grok 4.6effort: low— not measured—Est. cost / task—
    • Grok 4.6, effort: medium— not measured—Est. cost / task—
    • Grok 4.6, effort: highGrok 4.6 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.059Est. cost / task$0.0504Mean input tokens1,624Mean output tokens7,865Tests40IntervalNov 4 – 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.059Est. cost / task$0.0504
    • Grok 4.6, effort: xhigh— not measured—Est. cost / task—
    • Claude Opus 5effort: low— not measured—Est. cost / task—
    • Claude Opus 5, effort: medium— not measured—Est. cost / task—
    • Claude Opus 5, effort: highClaude Opus 5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.064Est. cost / task$0.0538Mean input tokens1,487Mean output tokens1,856Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.064Est. cost / task$0.0538
    • Claude Opus 5, effort: xhigh— not measured—Est. cost / task—
    • Claude Opus 5, effort: max— not measured—Est. cost / task—
    • Gemini 3.1 Pro Previeweffort: lowGemini 3.1 Pro Preview · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.07Est. cost / task$0.0595Mean input tokens999Mean output tokens4,789Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.07Est. cost / task$0.0595
    • Gemini 3.1 Pro Preview, effort: mediumGemini 3.1 Pro Preview · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.09Est. cost / task$0.0763Mean input tokens999Mean output tokens6,193Tests40IntervalOct 29 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.09Est. cost / task$0.0763
    • Gemini 3.1 Pro Preview, effort: highGemini 3.1 Pro Preview · effort: highRelative cost to Claude Opus 5.5 · max effort×0.12Est. cost / task$0.1016Mean input tokens999Mean output tokens8,300Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.12Est. cost / task$0.1016
    • Gemini 3.5 Flasheffort: minimal— not measured—Est. cost / task—
    • Gemini 3.5 Flash, effort: low— not measured—Est. cost / task—
    • Gemini 3.5 Flash, effort: mediumGemini 3.5 Flash · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.07Est. cost / task$0.0596Mean input tokens999Mean output tokens6,453Tests40IntervalOct 16 – Nov 17, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.07Est. cost / task$0.0596
    • Gemini 3.5 Flash, effort: high— not measured—Est. cost / task—
    • GPT-6 Astraeffort: lowGPT-6 Astra · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.087Est. cost / task$0.0735Mean input tokens970Mean output tokens1,276Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.087Est. cost / task$0.0735
    • GPT-6 Astra, effort: mediumGPT-6 Astra · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.091Est. cost / task$0.0772Mean input tokens970Mean output tokens1,350Tests40IntervalNov 11 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.091Est. cost / task$0.0772
    • GPT-6 Astra, effort: highGPT-6 Astra · effort: highRelative cost to Claude Opus 5.5 · max effort×0.11Est. cost / task$0.0951Mean input tokens970Mean output tokens1,709Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.11Est. cost / task$0.0951
    • GPT-6 Astra, effort: xhighGPT-6 Astra · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.15Est. cost / task$0.1298Mean input tokens970Mean output tokens2,402Tests40IntervalNov 4 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.15Est. cost / task$0.1298
    • GPT-6 Astra, effort: maxGPT-6 Astra · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.31Est. cost / task$0.2609Mean input tokens970Mean output tokens5,023Tests40IntervalNov 10 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.31Est. cost / task$0.2609
    • Claude Fable 5.2effort: low— not measured—Est. cost / task—
    • Claude Fable 5.2, effort: medium— not measured—Est. cost / task—
    • Claude Fable 5.2, effort: highClaude Fable 5.2 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.1Est. cost / task$0.0864Mean input tokens1,489Mean output tokens1,863Tests40IntervalNov 10 – 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.1Est. cost / task$0.0864
    • Claude Fable 5.2, effort: xhigh— not measured—Est. cost / task—
    • Claude Fable 5.2, effort: max— not measured—Est. cost / task—
    • Claude Fable 5.1effort: lowClaude Fable 5.1 · effort: lowRelative cost to Claude Opus 5.5 · max effort×0.12Est. cost / task$0.1000Mean input tokens1,489Mean output tokens1,702Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.12Est. cost / task$0.1000
    • Claude Fable 5.1, effort: mediumClaude Fable 5.1 · effort: mediumRelative cost to Claude Opus 5.5 · max effort×0.13Est. cost / task$0.1089Mean input tokens1,489Mean output tokens1,880Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.13Est. cost / task$0.1089
    • Claude Fable 5.1, effort: highClaude Fable 5.1 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.14Est. cost / task$0.1217Mean input tokens1,489Mean output tokens2,136Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.14Est. cost / task$0.1217
    • Claude Fable 5.1, effort: xhighClaude Fable 5.1 · effort: xhighRelative cost to Claude Opus 5.5 · max effort×0.26Est. cost / task$0.2190Mean input tokens1,489Mean output tokens4,083Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.26Est. cost / task$0.2190
    • Claude Fable 5.1, effort: maxClaude Fable 5.1 · effort: maxRelative cost to Claude Opus 5.5 · max effort×0.7Est. cost / task$0.5883Mean input tokens1,489Mean output tokens11,468Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.7Est. cost / task$0.5883
    • Claude Fable 5effort: low— not measured—Est. cost / task—
    • Claude Fable 5, effort: medium— not measured—Est. cost / task—
    • Claude Fable 5, effort: highClaude Fable 5 · effort: highRelative cost to Claude Opus 5.5 · max effort×0.12Est. cost / task$0.1027Mean input tokens1,487Mean output tokens1,756Tests40IntervalOct 21 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.12Est. cost / task$0.1027
    • Claude Fable 5, effort: xhigh— not measured—Est. cost / task—
    • Claude Fable 5, effort: max— not measured—Est. cost / task—
    • Muse Wonder 1.0no effort settingMuse Wonder 1.0 · no effort settingRelative cost to Claude Opus 5.5 · max effort×0.28Est. cost / task$0.2408Mean input tokens986Mean output tokens11,845Tests40IntervalOct 16 – Nov 18, 2026 UTCDollar estimate: mean input and output tokens at current standard rates. Cache discounts, price tiers and tools are excluded. Output includes reasoning tokens where reported.×0.28Est. cost / task$0.2408

Bar: from a model’s cheapest effort level to its dearest. Dot: one effort level. Ring: the toolbar’s effort level. Line: ×1.

Dot: one effort level. Line: ×1.

Average cost of our 800+ word writing test at current token prices. Your costs will vary by task. The scale uses equal steps for tenfold changes: ×0.01, ×0.1, ×1. Current models are tested at each reasoning effort; older versions at their default effort.

About the data

What does this page show?

Performance and reliability measured through MindsHub Inference, alongside current token prices, API aliases and fallback routes. The figures describe the models as served through MindsHub. They are not a ranking of every provider’s deployment of those models.

Where do the measurements come from?

We run automated API checks and writing tasks through MindsHub Inference. Complex task, writing speed and cost per task figures refresh every two hours, and the other measurements every six hours. Complex task performance and cost cover up to 35 days of tests. Each chart shows its measurement dates. If an update fails, we keep previously saved measurements with their original dates. Prices, aliases and fallback routes come from the same live catalog as our pricing page. A dash means no usable measurement is available.

How is Ping measured?

Every 10 minutes, checks from multiple regions ask for a one-word reply at the lowest reasoning effort. We time the first answer token, including any thinking before it. Ping measures basic responsiveness and tracks availability.

Sample prompt

Reply with the single word: ok
How is the Simple task measured?

Every six hours, the model receives a short briefing about a fictional market hall and is asked to describe a Saturday morning in at least 150 words. It must interpret the facts, decide how to present them and follow the writing constraints. We time the first answer token at the default reasoning effort, including thinking time.

Sample prompt, abridged; briefing omitted

Using only the supplied market-hall briefing, describe a typical Saturday morning in at least 150 words. Use plain paragraphs, with no headings or lists, and do not add facts.
How is the Complex task measured?

The model receives a longer briefing about the same fictional market hall and must turn it into a complete visitor’s guide of at least 800 words. It must analyze and organize the supplied facts, cover them in order and obey content and formatting constraints. We measure the first answer token, total response time and writing speed at each tested effort level. The first-answer and total times include thinking; writing speed starts when the answer begins. Repeated tests run over several weeks; each model’s details show the test count and interval. These are speed tests, not scores for reasoning quality.

Sample prompt, abridged; briefing omitted

Using only the supplied briefing, write a visitor’s guide of at least 800 words. Cover every part in the given order, in plain paragraphs with no headings, lists or tables. Do not add facts.
What are p50, p05 and p95?

P50 is the median: half of measurements are at or below it and half at or above it. P05 and P95 mark the lower and upper ends of the middle 90% of measurements. The leaderboard shows p50. Open Performance to explore the spread; Ping and Simple task have p50 and p95 only.

Does availability below 100% mean requests failed?

Not necessarily. A configured fallback route can still answer when the model you chose does not. Availability before fallback is the share of our automated checks answered by that model, including retries, over the selected period. Other checks were answered by a fallback or got no answer. These checks do not measure how often customer requests fail. Failures attributed to our monitoring network are excluded.

What does the intelligence score mean?

MindsHub’s read is a subjective estimate from 1 to 10 based on public benchmarks and rankings, not a precise measurement. Performance depends on your workload; test models on your own tasks.

For detailed benchmarks, refer to third party services such as Artificial Analysis, Epoch AI or similar.

How is cost per task compared?

The most expensive measured model and effort level is the 100% reference, updated automatically as results change. The leaderboard range covers a model’s measured effort levels; expand it for individual values. In Cost per task, choose another reference to compare against. Dollar estimates use mean input and output tokens at current standard rates, excluding cache discounts. Your bill depends on your requests.

What changes with reasoning effort?

Intelligence, Complex task timings and writing speed follow the selected effort. Ping always uses the lowest effort and Simple task the default. The cost range covers all measured levels. Views showing all effort levels disable the selector because every level is already visible.

What are fixed and rotating aliases?

An alias is the name sent in the API’s model field. Fixed aliases select a specific model version. Rotating aliases follow a model family as it updates. MindsHub Air and MindsHub Blaze are managed rotating aliases: MindsHub can change the model behind each name. The Aliases tab explains their purpose and lists the names you can use.

How do token prices, context limits and the free tier work?

Tokens are pieces of text. The context limit is how much input a model accepts in one request. Token prices are US dollars per million input or output tokens, from the same catalog as the pricing page. MindsHub Air is free within the free-tier limits; its listed token rates apply to usage beyond those limits. TypeSafe’s Jev is free. See Prices for cached-input, cache-write and other rates.

Which models are included?

The leaderboard covers chat models and their managed aliases. Other models lists decision and embedding models with their own response-time measurements and availability. See mindshub.ai/unified-inference#free-tier for the free tier and mindshub.ai/pricing for rates and credits.