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

Prices

Current rates in US dollars, the same as the pricing page. See pricing for free-tier terms and credits.

Showing 48 models

What the columns mean
Model

Current model rates. Separate rows appear when API names for the same model have different prices.

Aliases

Names you can send in the API’s model field. “Free tier” means free within fair use.

Host

The provider serving this model through MindsHub.

Input, per 1M

US dollars per million tokens you send.

Output, per 1M

US dollars per million tokens the model generates.

Cached input, per 1M

US dollars per million input tokens reused from the prompt cache.

Cache write, per 1M

US dollars per million tokens saved to the prompt cache.

Web search, per 1,000

US dollars per 1,000 web searches. Page fetches are shown separately when billed separately.

Long prompts

Rates for prompts above the listed token limit. These rates apply to the whole request.

A request whose prompt (uncached input, cached reads and cache writes together) is longer than this is priced entirely at the long-prompt rates; output does not count toward it.

Every price the live catalog sets, one row per model and price: input, output, cached input and cache write per 1M tokens, web search per 1,000, and long-prompt rates.

Model

Current model rates. Separate rows appear when API names for the same model have different prices.

Aliases

Aliases

Names you can send in the API’s model field. “Free tier” means free within fair use.

Host

Host

The provider serving this model through MindsHub.

Input, per 1M

US dollars per million tokens you send.

Output, per 1M

US dollars per million tokens the model generates.

Cached input, per 1M

US dollars per million input tokens reused from the prompt cache.

Cache write, per 1M

US dollars per million tokens saved to the prompt cache.

Long prompts

Long prompts

Rates for prompts above the listed token limit. These rates apply to the whole request.

A request whose prompt (uncached input, cached reads and cache writes together) is longer than this is priced entirely at the long-prompt rates; output does not count toward it.

MindsHub Air MindsHub Air is free within free-tier limits. The listed rates apply to usage beyond those limits. mindshub_air Free tier MindsHub $0.10 $0.50 $0.01 $0.125 $10 over 272,000 tokens$0.20 in · $0.75 outcached $0.02 · write $0.25
MindsHub Blaze mindshub_blaze Cerebras $0.99 $1.49 $0.99 No charge not set not set
GPT-OSS 120B (Fireworks) gpt-oss-fireworks Fireworks AI $0.15 $0.60 $0.02 $0.15 not set not set
Claude Sonnet 5.5 sonnet sonnet-5-5 Anthropic $2 $10 $0.20 $2.50 $10 not set
Claude Sonnet 5 sonnet-5 Anthropic $2 $10 $0.20 $2.50 $10 not set
Claude Opus 5.5 opus opus-5-5 Anthropic $4 $20 $0.20 $5 $10 not set
Claude Opus 5 opus-5 Anthropic $5 $25 $0.50 $6.25 $10 not set
Claude Fable 5.1 fable fable-5-1 Anthropic $10 $50 $0.25 $12.50 $10 not set
Claude Fable 5.2 fable-5-2 Anthropic $8 $40 $0.20 $10 $10 not set
Claude Fable 5 fable-5 Anthropic $10 $50 $1 $12.50 $10 not set
Claude Haiku 5.5 haiku haiku-5-5 Anthropic $0.10 $0.50 $0.01 $0.125 $10 over 100,000 tokens$0.50 in · $2.50 outcached $0.05 · write $0.625
Claude Haiku 4.5 haiku-4-5 Anthropic $1 $5 $0.10 $1.25 $10 not set
GPT-6 Astra gpt gpt-6-astra OpenAI $10 $50 $1 $12.50 $10 over 272,000 tokens$20 in · $75 outcached $2 · write $25
GPT 5.6 Sol gpt-5-6-sol OpenAI $5 $30 $0.50 $6.25 $10 over 272,000 tokens$10 in · $45 outcached $1 · write $12.50
GPT-6.1 Sol gpt-sol gpt-sol-6-1 OpenAI $2 $10 $0.10 $2.50 $10 over 272,000 tokens$4 in · $15 outcached $0.20 · write $5
GPT-6 Sol gpt-sol-6 OpenAI $2 $10 $0.20 $2.50 $10 over 272,000 tokens$4 in · $15 outcached $0.40 · write $5
GPT-6.1 Terra gpt-terra gpt-terra-6-1 OpenAI $1 $5 $0.05 $1.25 $10 over 272,000 tokens$2 in · $7.50 outcached $0.10 · write $2.50
GPT 5.6 Terra gpt-terra-5-6 OpenAI $2 $12 $0.20 $2.50 $10 over 272,000 tokens$4 in · $18 outcached $0.40 · write $5
GPT-6 Luna gpt-luna gpt-luna-6 OpenAI $0.10 $0.50 $0.01 $0.125 $10 over 272,000 tokens$0.20 in · $0.75 outcached $0.02 · write $0.25
GPT 5.6 Luna gpt-luna-5-6 OpenAI $0.20 $1.20 $0.02 $0.25 $10 over 272,000 tokens$0.40 in · $1.80 outcached $0.04 · write $0.50
GPT 5.3 Codex gpt-codex gpt-codex-5-3 OpenAI $1.75 $14 $0.18 $1.75 $10 not set
GPT 5.4 Mini gpt-mini gpt-mini-5-4 OpenAI $0.75 $4.50 $0.08 $0.75 $10 not set
GPT 5.4 Nano gpt-nano gpt-nano-5-4 OpenAI $0.20 $1.25 $0.02 $0.20 $10 not set
Gemini 3.1 Pro Preview gemini gemini-3-1-pro Google $2 $12 $0.20 No charge $14 over 200,000 tokens$4 in · $18 outcached $0.40 · write No charge
Gemini 4 Argon gemini-argon gemini-argon-4 Google $2 $10 $0.10 No charge $14 not set
Gemini 3.8 Flash gemini-flash gemini-flash-3-8 Google $0.75 $3.75 $0.08 No charge $14 not set
Gemini 3.7 Flash gemini-flash-3-7 Google $0.75 $3.75 $0.08 No charge $14 not set
Gemini 3.6 Flash gemini-flash-3-6 Google $0.75 $3.75 $0.08 No charge $14 not set
Gemini 3.5 Flash gemini-flash-3-5 Google $1.50 $9 $0.15 No charge $14 not set
Gemini 3 Flash Preview gemini-flash-3 Google $0.50 $3 $0.05 No charge $14 not set
Gemini 3.1 Flash-Lite gemini-flash-lite gemini-flash-lite-3-1 Google $0.25 $1.50 $0.03 No charge $14 not set
Kimi K3 kimi kimi-k3 Fireworks AI $3 $15 $0.30 $3 searches $7page fetches $1 not set
Kimi K3.1 Pro kimi-pro kimi-k3-1-pro Moonshot AI $2 $10 $0.20 No charge searches $7page fetches $1 not set
DeepSeek V4.1 Flash 1017 deepseek-v4-flash deepseek-v4-1-flash-1017 Fireworks AI $0.30 $1.20 $0.006 $0.30 searches $7page fetches $1 not set
DeepSeek V4.1 Flash deepseek-v4-1-flash Fireworks AI $0.30 $1.20 $0.006 $0.30 searches $7page fetches $1 not set
Qwen3.8-2.4T-A95B qwen qwen-3-8-a95b Fireworks AI $2 $6 $0.25 $2 searches $7page fetches $1 not set
GLM 5.3 glm glm-5-3 Fireworks AI $1.40 $4.40 $0.26 $1.40 searches $7page fetches $1 not set
GLM 5.3 Flash glm-5-3-flash glm-flash-5-3 Fireworks AI $0.15 $0.50 $0.03 $0.15 searches $7page fetches $1 not set
Muse Spark 1.3 muse-spark muse-spark-1-3 Meta $1.25 $4.25 $0.15 $1.25 searches $7page fetches $1 not set
Muse Spark 1.2 muse-spark-1-2 Meta $1.25 $4.25 $0.15 $1.25 searches $7page fetches $1 not set
Muse Spark 1.1 muse-spark-1-1 Meta $1.25 $4.25 $0.15 $1.25 searches $7page fetches $1 not set
Muse Wonder 1.0 muse-wonder muse-wonder-1-0 Meta $4 $20 $0.20 $4 searches $7page fetches $1 not set
Muse Ember 1.0 muse-ember muse-ember-1-0 Meta $0.10 $0.50 $0.004 $0.10 searches $7page fetches $1 over 200,000 tokens$0.30 in · $1.50 outcached $0.012 · write $0.30
Grok 4.7 grok grok-4-7 SpaceXAI $2 $6 $0.50 $2 $5 over 199,999 tokens$4 in · $12 outcached $1 · write $4
Grok 4.6 grok-4-6 SpaceXAI $2 $6 $0.50 $2 $5 over 199,999 tokens$4 in · $12 outcached $1 · write $4
Grok 4.5 grok-4-5 SpaceXAI $2 $6 $0.30 $2 $5 over 199,999 tokens$4 in · $12 outcached $0.60 · write $4
Text Embedding 3 (small) Embedding model embed-small embed-3-small OpenAI $0.02 not set not set not set not set not set
Jev 1.13.0 Decision model jev jev-1.13.0 TypeSafe $0 $0 not set not set not set not set

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.