The AI arena is free today

Open Superagent
API Provider4 active models1 organization

Mistral AI: API pricing, speed & models

Mistral AI hosts 4 active AI models, with input pricing from $0.15 per 1M tokens, with median throughput of 99 characters/sec, and P95 time to first token of 8.64s, with 100% success rate over 7 days. Compare Mistral AI's API speed, pricing, and reliability against other inference providers.

First-party only
Median throughput99 c/s
Median TTFT7.54s
P95 TTFT8.64s
Success rate (7d)100%
From$0.15 /M tok

Catalog

Type
Price
7 models
Model
At a glance

Mistral AIpricing, performance & catalog

The citable facts about Mistral AI's 4 models — sourced from provider APIs and refreshed continuously.

Lowest price
Mistral Small 4 at $0.150 per 1M input tokens
Highest median throughput
Mistral Large 3 (675B Instruct 2512) at 99 chars/s
Lowest median TTFT
Mistral Large 3 (675B Instruct 2512) at 1.35s
Largest context
Mistral Large 3 (675B Instruct 2512) at 262K tokens
Catalog
4 active models from 1 organization

FAQ

Common questions about Mistral AI.

What is Mistral AI?

Mistral AI is an API provider that hosts large language models. Active models: 4; From (input): $0.15 / 1M tok; Median throughput: 99 c/s; Median TTFT: 7.54s; Success rate (7d): 100%.

How many models does Mistral AI offer?

Mistral AI currently serves 4 active models out of 18 historical offerings on LLM Stats.

What is Mistral AI's API pricing?

Mistral AI input pricing starts from $0.15 per 1M tokens, with the most expensive offering at $1.5 per 1M tokens. See the Pricing tab above for the full per-model breakdown.

How fast is Mistral AI?

Mistral AI delivers a median throughput of 99 characters per second and P95 time to first token of 8.64s across its model catalog. See the Models tab for per-model throughput and TTFT breakdowns.

Is Mistral AI reliable?

Mistral AI has a 100% success rate across 3 API calls in the last 7 days, with a 0% error rate.

Does Mistral AI support function calling?

Yes. 6 of 4 models on Mistral AI support function calling (tool use). The Capabilities tab lists which specific models accept tool definitions.

Does Mistral AI support JSON mode and structured output?

Yes. 6 of 4 models on Mistral AI support structured output (JSON mode / schema-constrained generation). The Capabilities tab shows which specific models accept response_format or json_schema parameters.

Does Mistral AI offer batch inference?

Yes. 6 of 4 models on Mistral AI support batch inference for cheaper, asynchronous workloads.

Does Mistral AI support multimodal models?

Yes. Mistral AI's catalog includes 3 vision-capable models. See the Models and Capabilities tabs for the full per-model breakdown.

Whose models does Mistral AI host?

Mistral AI hosts models from Mistral AI. See the Models tab for the full catalog grouped by creator.

How do I start using Mistral AI?

Sign up at https://mistral.ai to get an API key, then call Mistral AI's API directly from your application. Most clients work out of the box by pointing the OpenAI SDK at Mistral AI's base URL with your key. Use the Models and Pricing tabs above to pick the right model for your latency, cost, and context-window requirements.