Magistral Small

Tool callingReasoning

Website

Open Mistral reasoning model for transparent step-by-step problem solving

Context: 131K
Released: 1y ago
Parameters: —

Sources:models.dev

Benchmarks

DeepSWE is the primary coding benchmark where an exact model evaluation exists. Other sourced evaluations appear below; missing scores are not zero.

Not evaluated in DeepSWE 1.1 as of 2026-09-22. Results are only shown for exact model versions in the official dataset. Official leaderboard.

Other published benchmarks

Evaluation	Reported score and source
ARC-AGI-2	0.0% Third-party evaluator · magistral-small-2506 External evaluation via Epoch AI Original-source terms apply
GPQA diamond	56.1% Epoch AI · magistral-small-2506 Epoch AI evaluation
OTIS Mock AIME 2024-2025	30.0% Epoch AI · magistral-small-2506 Epoch AI evaluation

Benchmark compilation: Epoch AI, Capabilities & benchmarking (CC BY attribution). Scores are converted to percentages and model names matched to the Vertical Tabs catalog. Individual evaluations are credited to their original sources. External results may carry separate licences and evaluation configurations; do not interpret scores from different harnesses as directly comparable.

Specifications

Developer: Mistral AI
Model family: Magistral
Type: Language model
Predecessor: —
Successor: —
Released: June 10, 2025
Model ID: mistral/magistral-small-2506
Openness: Open weights
License: Apache 2.0
Parameters: —
Input modalities: text
Output modalities: text
Modality: text->text
Max output tokens: 8K
Hugging Face: —
Source freshness: Metadata synced Oct 8, 2026, 10:17 PM UTC

Additional source evidence

Source model: mistral/magistral-small-2506
Source family: magistral
Last updated: 2025-06-10
Model context: 131K
Model output limit: 8K
Open weights: Yes
File attachments: No
Reasoning: Yes
Tool calling: Yes
Temperature control: Yes

Downloadable weights

Hugging Face

Complete source record (JSON)

{
  "id": "mistral/magistral-small-2506",
  "name": "Magistral Small",
  "description": "Open Mistral reasoning model for transparent step-by-step problem solving",
  "family": "magistral",
  "last_updated": "2025-06-10",
  "attachment": false,
  "reasoning": true,
  "tool_call": true,
  "temperature": true,
  "modalities": {
    "input": [
      "text"
    ],
    "output": [
      "text"
    ]
  },
  "limit": {
    "context": 131072,
    "output": 8192
  },
  "release_date": "2025-06-10",
  "open_weights": true,
  "license": "Apache 2.0",
  "weights": [
    {
      "label": "Hugging Face",
      "url": "https://huggingface.co/mistralai/Magistral-Small-2506"
    }
  ]
}

Infrastructure

Provider availability

All providers

Provider endpoint data will appear after this model's first OpenRouter endpoint sync.

Deployment graph

Can I run this?

Vertical Tabs only shows compatibility when it has structured evidence tying this exact model to hardware, runtime, quantization, memory use, and observed performance.

Browse hardware