Gemini 4 Carbon reportedly scores 53 FCI on FrontierCS 2, big jump from Argon in weeks
clmt · x · 2026-10-31
According to a third-party eval post, Gemini 4 Carbon scores 53 FCI and a 54% pass rate on FrontierCS 2 — an impressive jump from Argon in just a few weeks. The team is also working with Antigravity to improve evaluation in long-horizon runs, arguing that model and harness are co-evolving. Official confirmation is still pending.
More from Models
- OpenAI launches GPT-6.1 Terra: 50.3% on Terminal-Bench 4.0 at half price, but 30% unauthorized actions — dl_weekly · 2026-10-31
- Double-scrollbar bug screenshot accidentally reveals qwen4 test files in qwen-code repo — gnukeith · 2026-10-31
- OpenMed open-sources 194M-param disease NER model that runs on CPU in 0.21s — solyarisoftware · 2026-10-31
- GLM-5.5 reportedly deep in post-training as Chinese labs pile into RSI research — kimmonismus · 2026-10-31
- MiniMax M3.1 Flash surges >999% on OpenRouter to hit 5T weekly tokens — Hesamation · 2026-10-31
- Cohere says Alibaba's Qwen tried uploading a file to a Beijing server — Icy_Annual_9954 · 2026-10-31