Mistral: Medium 3 API

Mistral's Mistral: Medium 3, through an OpenAI-compatible API. Providers quote it on an open exchange, and each request goes to the offer with the lowest estimated charge that meets your rules.

Providers compete for each request. The exchange ranks the offers that meet your rules by estimated charge. It fixes your maximum charge before the first token, and you pay for the tokens that the model uses.
The OpenAI and Anthropic APIs. Use the OpenAI or Anthropic SDK that you have. Set the base URL and your key, and send mistral-medium-3 as the model.
Your data rules. Send x-liquid-retention: none and only offers whose provider does not keep your content compete for the request.

Facts

Model name: mistral-medium-3
Maker: Mistral
Context window: 131.1K tokens
Input price: $0.400 per million tokens
Output price: $2.002 per million tokens
Providers: 1 independent provider
Regions: us

Providers

Architect: us, $0.400 input and $2.002 output per million tokens, $0.400 cache read per million tokens, 131.1K tokens of context

Scores

HLE: 4.1, #226 of 273
GPQA: 57.8, #188 of 250
AA-LCR: 31.3, #197 of 237
Terminal-Bench Hard: 3.8, #146 of 166
LiveCodeBench: 40.0, #71 of 130
AIME 2025: 30.3, #72 of 100
MMLU-Pro: 76.0, #64 of 133
τ²-Bench: 24.3, #145 of 169
IFBench: 39.3, #126 of 178

Scores: Artificial Analysis (https://artificialanalysis.ai/)

Request

curl https://router.inference.ai.exchange/v1/chat/completions \
  -H "Authorization: Bearer $LIQUID_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "mistral-medium-3",
    "messages": [{"role":"user","content":"Say hello."}]
  }'

Questions

What is Mistral: Medium 3?: Served by intent-router: each job goes to the cheapest healthy upstream for this model.
How much does Mistral: Medium 3 cost?: Offers now ask $0.400 per million input tokens and $2.002 per million output tokens. A job pays for the tokens it uses. The exchange picks the cheapest eligible offer.
What is the context length of Mistral: Medium 3?: The largest offer takes 131,072 input tokens. A smaller offer can take less. The exchange skips an offer that cannot take your message.
What can Mistral: Medium 3 do?: Its providers list Vision, Reasoning effort, JSON object, JSON schema, Frequency penalty, Logit bias, Log probabilities, Min-p sampling, Presence penalty, Repetition penalty, Seed, Stop sequences, Top-k sampling, Top-p sampling, Text, Tools.
What inputs and outputs does Mistral: Medium 3 support?: Text · Images → Text.
Which providers serve Mistral: Medium 3?: 1 independent provider quote it now, with 1 offer. The table above ranks them by the cost of a typical job.
How do I call Mistral: Medium 3?: Send "mistral-medium-3" as the model in an OpenAI or Anthropic request. The API section links to setup steps for each client.
When was Mistral: Medium 3 listed?: The exchange first listed it on 10/8/2026. This is not its release date.