Devstral: Small 2505 API

Mistral's Devstral: Small 2505, 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 devstral-small-2505 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: devstral-small-2505
Maker: Mistral
Context window: 32.8K tokens
Input price: $0.060 per million tokens
Output price: $0.060 per million tokens
Providers: 1 independent provider
Regions: us

Providers

Architect: us, $0.060 input and $0.060 output per million tokens, $0.060 cache read per million tokens, 32.8K tokens of context

Scores

HLE: 4.0, #240 of 273
GPQA: 43.4, #222 of 250
AA-LCR: 32.0, #195 of 237
Terminal-Bench Hard: 6.1, #140 of 166
LiveCodeBench: 25.8, #104 of 130
MMLU-Pro: 63.2, #115 of 133
τ²-Bench: 38.0, #117 of 169
IFBench: 31.6, #158 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": "devstral-small-2505",
    "messages": [{"role":"user","content":"Say hello."}]
  }'

Questions

What is Devstral: Small 2505?: Devstral: Small 2505 is a model from Mistral.
How much does Devstral: Small 2505 cost?: Offers now ask $0.060 per million input tokens and $0.060 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 Devstral: Small 2505?: The largest offer takes 32,768 input tokens. A smaller offer can take less. The exchange skips an offer that cannot take your message.
What can Devstral: Small 2505 do?: Its providers list 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 Devstral: Small 2505 support?: Text → Text.
Which providers serve Devstral: Small 2505?: 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 Devstral: Small 2505?: Send "devstral-small-2505" as the model in an OpenAI or Anthropic request. The API section links to setup steps for each client.
When was Devstral: Small 2505 listed?: The exchange first listed it on 10/10/2026. This is not its release date.