Baltor Get started

Models

Mistral Small 4

By Mistral. 119.4 billion parameters, a context window of 262,144 tokens and the licence apache-2.0.

Facts

Released
2026-01-23 the day the repository was first published on Hugging Face Hugging Face, read
Licence
apache-2.0 the licence the model card declares Hugging Face, read
Open weights
Yes the weights are published in this Hugging Face repository Hugging Face, read
Parameters
119.4 billion counted from the safetensors weight files Hugging Face, read
Active parameters
Unknown
Knowledge cutoff
Unknown
Context window
  • 262,144 tokens the context length OpenRouter lists OpenRouter, read
Longest output
  • 209,715 tokens the largest output OpenRouter's first provider allows OpenRouter, read
Tool calling
  • Yes OpenRouter lists tools among the supported parameters OpenRouter, read
Structured output
  • Yes OpenRouter lists structured outputs among the supported parameters OpenRouter, read
Reasoning controls
  • Yes OpenRouter lists reasoning controls, with the efforts high, none OpenRouter, read
Inputs and outputs
text, image in, text out OpenRouter, read
Good for
  • Reasoning: OpenRouter lists reasoning controls for it OpenRouter, read
  • Vision: OpenRouter lists images among its inputs OpenRouter, read

Prices

Prices in US dollars per million tokens, as each source lists them
ProviderInputOutputContextAs ofSource
Nvidia direct0 USD0 USD128,000 older than 30 days; check the providermodels.dev, read
Mistral through OpenRouter0.150 USD0.600 USD262,144OpenRouter endpoints, read
Mistral through OpenRouter0.150 USD0.600 USD262,144OpenRouter endpoints, read
Mistral through OpenRouter0.150 USD0.600 USD262,144OpenRouter endpoints, read
Mistral through OpenRouter0.150 USD0.600 USD262,144OpenRouter endpoints, read
Pioneer direct0.150 USD0.600 USD32,000 older than 30 days; check the providermodels.dev, read
Mistral through OpenRouter0.165 USD0.660 USD262,144OpenRouter endpoints, read
Mistral through OpenRouter0.165 USD0.660 USD262,144OpenRouter endpoints, read
Infomaniak direct0.250 USD0.930 USD256,000 older than 30 days; check the providermodels.dev, read
NanoGPT direct0.400 USD1.40 USD262,144 older than 30 days; check the providermodels.dev, read

Run it on your own hardware

Weights are the sizes of the files a source lists, or an estimate from the parameter count where none does. Memory is an estimate: weights plus KV cache plus 512 MiB and 5 percent of the weights for runtime buffers. Check it against your hardware.

QuantizationWeightsMemory at 8,192 tokensMemory at 32,768 tokens
F16 estimate222 GiBUnknown: no KV cache numbersUnknown: no KV cache numbers
Q8_0 estimate118 GiBUnknown: no KV cache numbersUnknown: no KV cache numbers
Q4_0 estimate62.6 GiBUnknown: no KV cache numbersUnknown: no KV cache numbers

Start it with a local runtime

  • vLLM
    vllm serve mistralai/Mistral-Small-4-119B-2603
  • SGLang
    sglang serve --model-path mistralai/Mistral-Small-4-119B-2603 --port 30000

Use it from your harness

Set up for Infomaniak with the model mistralai/Mistral-Small-4-119B-2603. Each endpoint page has the same setup for its own address.

OpenCode

Put this in opencode.json in your project folder:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "infomaniak": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Infomaniak",
      "options": {
        "baseURL": "https://api.infomaniak.com/2/ai/${INFOMANIAK_PRODUCT_ID}/openai/v1",
        "apiKey": "{env:INFOMANIAK_API_KEY}"
      },
      "models": {
        "mistralai/Mistral-Small-4-119B-2603": {
          "name": "Mistral Small 4",
          "limit": {
            "context": 262144,
            "output": 209715
          }
        }
      }
    }
  }
}
  • OpenCode reads any OpenAI-compatible address through the @ai-sdk/openai-compatible package, and an address that speaks the Responses API through @ai-sdk/openai.

From OpenCode documentation, read .

Pi

Put this in ~/.pi/agent/models.json:

{
  "providers": {
    "infomaniak": {
      "baseUrl": "https://api.infomaniak.com/2/ai/${INFOMANIAK_PRODUCT_ID}/openai/v1",
      "api": "openai-completions",
      "apiKey": "$INFOMANIAK_API_KEY",
      "models": [
        {
          "id": "mistralai/Mistral-Small-4-119B-2603"
        }
      ]
    }
  }
}
  • The apiKey field can name an environment variable as $NAME.

From Pi documentation, read .

Codex

Codex speaks only the Responses API, and Infomaniak documents no Responses address. A gateway that offers one can sit in between.

  • Codex speaks the Responses API only: responses is the one supported wire API of a custom provider. Ollama and LM Studio are built in and start with --oss.

From Codex documentation, read .

Claude Code

Claude Code sends Anthropic Messages requests, and Infomaniak documents no such address. A gateway that translates to that API can sit in between.

  • Claude Code sends Anthropic Messages requests to ANTHROPIC_BASE_URL. Anthropic says it does not support routing Claude Code to models other than Claude through any gateway, so some features may not work with another model.

From Claude Code documentation, read .

Other ways to reach it

  • NanoGPT, model mistralai/mistral-small-4-119b-2603
  • Nvidia, model mistralai/mistral-small-4-119b-2603
  • Pioneer, model mistralai/Mistral-Small-4-119B-2603
  • OpenRouter, model mistralai/mistral-small-2603

Published results

These are results other people published. Baltor did not run them and does not rank models by them.