Mistral Large 4 Scorecard: 49.8% Coding Agent Index, 93.3% Attack Resistance on Lakera B3
_AndrewZhao · x · 2026-10-29
An aggregator post summarizes the new open-weight 1T-parameter Mistral Large 4 (Le Chonk): 49.8% on the Coding Agent Index, 59.9% on AutomationBench, 93.3% attack resistance on Lakera's B3 security benchmark, and 1393 Elo on AA-Briefcase — billed as the best open-weight model from a Western lab. Charts show it leading FrontierCode 2.0 at 19.7% over DeepSeek V4.1 Flash (19.3%), GLM-5.3 (17.8%), Kimi K3 (17.3%), and Beam 1.0 (2.9%), with strong token efficiency on Artificial Analysis; blind human preference wins 64% vs GLM-5.3/Kimi K3 but only 22% vs Claude Opus 5.5.
More from Models
- Codex doubles usage limits for all Pro tiers after user backlash — pvncher · 2026-10-30
- Dream raises $17M seed from a16z to build spatial-reasoning AI for AEC — a16z · 2026-10-30
- Investor bets speech-to-speech will be massive, praises Phonic's model — graceisford · 2026-10-30
- User Praises OpenAI for Doubling Pro Usage and Including API in Pro Sub — Angaisb_ · 2026-10-30
- OpenAI doubles usage limits for Pro tiers and Business Premium, effective immediately — Polymarket · 2026-10-30
- Gemini 4 Argon tops 4 robotics tasks on RLE-Bench, ranks #2 behind GPT-6 Astra — daibond_alpha · 2026-10-30