# The Decoder — Mistral Large 4 is Europe's trillion-parameter answer to US models that refuse security work

- Company: The Decoder (the-decoder.com)
- Announced: 2026-10-06T14:18:08+00:00
- Category: new-model
- Coverage: not counted
- Announcement: yes
- Group: models
- Source: https://the-decoder.com/mistral-large-4-is-said-to-be-the-most-powerful-open-ai-model-from-europe-and-the-u-s/
- Record: https://forck.live/items/16217-mistral-large-4-is-europe-s-trillion-parameter-answer-to-us-models-that-refuse
- Subject: Reporting
- Open weights: yes
- Models affected: Mistral Large 4, Mistral Large 3, Mistral Medium 3.5, GLM-5.2, Claude Opus 5.5, GPT-6 Astra, Deepseek V4 Pro, Qwen3.8 Max, GLM-5.3, Kimi K3, Claude Opus 5

Mistral released a public preview of Mistral Large 4, a one-trillion-parameter multimodal model trained in Europe, with weights expected at the end of October. The model scores 38 points in the Artificial Analysis Intelligence Index, up from 9 for its predecessor, but trails Claude Opus 5.5 at 58. Mistral emphasizes cybersecurity strengths, including an 82 percent score on a vulnerability reproduction-and-patch test, which it attributes partly to rivals refusing such tasks.

## Evidence

Verbatim from https://the-decoder.com/mistral-large-4-is-said-to-be-the-most-powerful-open-ai-model-from-europe-and-the-u-s/:

> ML4 hits 82 percent, the highest score of any model. The reason is unusual, though. According to Mistral, Claude Opus 5.5 and GPT-6 Astra score near zero because they refuse the task entirely.

---

Record: https://forck.live/items/16217-mistral-large-4-is-europe-s-trillion-parameter-answer-to-us-models-that-refuse
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md