Post
29 Oct 2026
Mistral Large 4: 1T-parameter open weights, modified MIT licence
Mistral released the open weights of Mistral Large 4 on 29 October 2026: a ~1-trillion-parameter Mixture-of-Experts model (~52B active, 1M-token context, text + image input) downloadable from Hugging Face. Mistral reports 48% on the Artificial Analysis Coding Agent Index (tied with GLM-5.3 while using 47% fewer tokens per task) and 51% on the Artificial Analysis Cyber Index. The Mistral Ecosystem Open-Weight License is a modified MIT licence: free to use, modify, fine-tune and commercialize, but businesses operating a Model-as-a-Service offering need a separate commercial licence. API price $0.70/$2.10 per million input/output tokens. Benchmarks are Mistral-reported.
02ModelsGemini 4 Carbon: Google's new #1 AI model tops cyber benchmarks
Google announced Gemini 4 Carbon on 28 October 2026, its new frontier model. Google says it ranks #1 on the Artificial Analysis Intelligence Index v5.0 with a score of 59, and reports state-of-the-art cybersecurity results: 88.2% on SEC-bench Pro (finding vulnerabilities), 71.0% on CWE-bench (patching) and 54.8% on ExploitGym. Google says its teams have seen Carbon run for days on long, multi-step engineering tasks. Access starts with a subset of cyber defenders; developers, businesses and consumers come later. Introductory pricing is $2 per million input tokens and $10 per million output tokens ($4 / $20 after the introductory period). All benchmarks are Google-reported.
blog.google · blog.google ↗ (opens in a new tab)