Summary
According to a post on X by @getsuperintel, Mistral released the weights of Large 4 on Oct 29. The 1-trillion-parameter multimodal model had entered public preview earlier in the month, and three more weeks of reinforcement learning raised its Terminal-Bench 4.0 score by 13 points to 41%. The post says Large 4 leads open-weight models on Cognition's FrontierCode 2.0 at 19.7% and ties GLM-5.3 at 48% on Artificial Analysis's Coding Agent Index while using 47% fewer tokens. It also lists pricing of $0.70 input and $2.10 output per million tokens.
AI-assisted summary based on the listed source.
What happened
Mistral released the weights of Large 4 on Oct 29, the 1T-parameter multimodal model that entered public preview earlier this month; three more weeks of RL lifted Terminal-Bench 4.0 by 13 points to 41%. It leads open-weight models on Cognition's FrontierCode 2.0 at 19.7% and ties GLM-5.3 at 48% on Artificial...
Why it matters
Developers and companies evaluating open-weight models for coding agents may find these benchmark claims relevant. The figures come from a social post, so teams weighing self-hosting or API use should check them against the official release and independent evaluations. Watch for third-party benchmark results from Artificial Analysis and Cognition.
Signal Intelligence
Signal Strength 92%
Technical label WATCH
Public Interest 56
Category OPEN SOURCE
Reader Depth TECHNICAL
Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.
Public Interest components
Recognizable Entity Score 73
Practical Impact Score 44
Novelty Interest Score 80
Consequence Score 34
Curiosity Score 16
Shareability Score 66
Why this is here
VQV surfaced this signal because it is recent, relevant to Open Source LLMs, connected to Scobleizer News Pulse (X).
