Mistral AI

Mistral Large 3 2512

Tool callingStructured outputImage input
Mistral Large 3 2512 launch post cover

Mistral's largest general model for enterprise agents, coding, and multilingual reasoning

Context
262K
Released
10mo ago
Parameters
—
Input / 1M tokens
$0.50
Output / 1M tokens
$1.50
Cached input / 1M tokens
$0.05

Sources:models.devOpenRouter

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.

No benchmark results have been matched to this model yet.

Specifications

Developer
Mistral AI
Model family
Mistral Large
Type
Language model
Predecessor
—
Successor
—
Released
December 2, 2025
Model ID
mistralai/mistral-large-2512
Openness
Open weights
License
—
Parameters
—
Input modalities
text, image, file
Output modalities
text
Modality
text+image+file->text
Tokenizer
Mistral
Max output tokens
210K
Knowledge cutoff
2024-11
Hugging Face
—
Source freshness
Metadata synced Oct 8, 2026, 10:17 PM UTC
Additional source evidence
Source model
mistral/mistral-large-2512
Source family
mistral-large
Last updated
2025-12-02
Knowledge cutoff
2024-11
Model context
262K
Model output limit
262K
Open weights
Yes
File attachments
Yes
Reasoning
No
Tool calling
Yes
Temperature control
Yes

Downloadable weights

Published benchmarks

BenchmarkScoreEvaluation detailsEvidence
Artificial Analysis Coding Index22.7 · index2026-06-04Source
SciCode36.2 · percent correct2026-06-04Source
Terminal-Bench Hard15.9 · success rate2026-06-04Source
Complete source record (JSON)
{
  "id": "mistral/mistral-large-2512",
  "name": "Mistral Large 3",
  "description": "Mistral's largest general model for enterprise agents, coding, and multilingual reasoning",
  "family": "mistral-large",
  "last_updated": "2025-12-02",
  "knowledge": "2024-11",
  "attachment": true,
  "reasoning": false,
  "tool_call": true,
  "temperature": true,
  "modalities": {
    "input": [
      "text",
      "image"
    ],
    "output": [
      "text"
    ]
  },
  "limit": {
    "context": 262144,
    "output": 262144
  },
  "benchmarks": [
    {
      "name": "Artificial Analysis Coding Index",
      "score": 22.7,
      "source": "https://openrouter.ai/mistralai/mistral-large-2512/benchmarks",
      "metric": "index",
      "date": "2026-06-04"
    },
    {
      "name": "SciCode",
      "score": 36.2,
      "source": "https://openrouter.ai/mistralai/mistral-large-2512/benchmarks",
      "metric": "percent correct",
      "date": "2026-06-04"
    },
    {
      "name": "Terminal-Bench Hard",
      "score": 15.9,
      "source": "https://openrouter.ai/mistralai/mistral-large-2512/benchmarks",
      "metric": "success rate",
      "date": "2026-06-04"
    }
  ],
  "release_date": "2025-12-02",
  "open_weights": true,
  "weights": [
    {
      "label": "Hugging Face",
      "url": "https://huggingface.co/mistralai/Mistral-Large-3-675B-Instruct-2512"
    }
  ]
}

Infrastructure

Provider availability

All providers
ProviderEndpointContextInput /M tokensOutput /M tokensCached input /M tokensCache write /M tokensUptime 1dData handling
Mistralmistral/zdr262K$0.50$1.50$0.05—99.86%
ZDR: Yes

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Mistralmistral/zdrBatch pricing262K$0.50$1.50$0.05—99.86%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Mistralmistral/eu262K$0.55$1.65$0.06—100.00%
ZDR: Yes

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Mistralmistral/euBatch pricing262K$0.55$1.65$0.06—100.00%
ZDR: No

Training: Unknown

Prompt logging: Unknown

Retention: Unknown

Checked 2026-10-15

Provider data:OpenRouter

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