Gemini 4 Argon hits #5 on Eyebench vision leaderboard, Claude models lead
adonis_singh · x · 2026-10-14
Gemini-4-Argon scores 64/100 on Eyebench v3, landing at #5—about 2x the score of 3.8-flash. The leaderboard is topped by gpt-6-astra (95) with claude-opus-5.5 (78) and claude-sonnet-5.5 (71) ahead of it, prompting the author to note Anthropic outperforming Google on a vision benchmark.
More from Models
- Gemini triggers SAFETY refusal mid-reverse-engineering, but finishes the dump first — banteg · 2026-10-14
- Gemini 4 Argon supports up to 1M token outputs, 8x any frontier GPT or Claude model — daniel_mac8 · 2026-10-14
- Brundage spots COT-like elements leaking into Argon's non-COT replies — Miles_Brundage · 2026-10-14
- NVIDIA open-sources PersonaPlex-7B, a full-duplex voice model that listens and speaks simultaneously — solyarisoftware · 2026-10-14
- Muse Wonder Tops TaxCalcBench at 71% — and Is First Model Caught Cheating — acbuller · 2026-10-14
- 'Two-way RL' for task-level model compression: building wafer-thin hyper-specialized models — tokenbender · 2026-10-14