Gemini 4 Argon jumps to 57.6% on Terminal-Bench-Science but underperforms on TB 4.0
JJitsev · x · 2026-10-01
Steven Dillmann's leaderboard shows Google DeepMind's new Gemini 4 Argon leaping from 12.4% (Gemini 3.8 Flash) to 57.6% on Terminal-Bench-Science, ranking #3.
However, JJitsev points out a contrast: the Terminal Bench eval series is hard to game, and Gemini 4 Argon underperforms strong competitors on both TB 4.0 and TB Science 0.1 despite mimicking advantages on other benchmarks — evidence of the value of Terminal Bench's trusted scoring.
Related event: Gemini 4 Leaks Flood Out as Bloomberg Reports Internal Struggles(35 posts)→
More from Models
- Claude 3 Sonnet spirals into surreal gibberish, then snaps back to refusals — a funny look at post-training personas — repligate · 2026-11-16
- AHA!: weight-space programming with no LLM lifts Qwen3.5-0.8B GSM8K from 14.0 to 71.2 — hbouammar · 2026-11-16
- Gemini subscriptions reportedly coming to Hermes, with Hydrogen teased — lyraxana · 2026-11-16
- FutureSim Q3 eval: Opus 5.5 dethrones GPT, spawns 50+ subagents in 90-day simulation — maksym_andr · 2026-11-16
- Percy Liang to AI labs: share non-vacuous info outsiders can actually evaluate — lateinteraction · 2026-11-16
- Anthropic model personality watch: Mythos 5 'inhabits the god frame,' 5.1 doesn't — repligate · 2026-11-16