Rumor: Insanely fast Gemini 4 on LMArena may be Gemini 4 Flash at ~300tps
lyraxana · x · 2026-10-27
A leak from @lyraxana suggests Gemini 4 Flash runs at around 300 tokens per second, notably fast. The tip: if you test Gemini 4 on LMArena and it feels insanely fast, it's likely the next-gen "hydrogen bomb" Flash model. Unverified, but a useful tell for spotting the new model testing anonymously on the arena.
More from Models
- GPT-6.1 Terra hits 78.5 on Browser Use Benchmark, near Sol at 35% less cost — sandersted · 2026-10-27
- Microsoft's MAI-Image claims #2 spot in image editing, beating Google, Meta and xAI — mustafasuleyman · 2026-10-27
- Reddit asks: which 3-8B small model handles real-time EN/CN translation well? — Loose_Comparison368 · 2026-10-27
- Insider says actual V4.1 Pro will be similarly built, current version not materially stronger — teortaxesTex · 2026-10-27
- Microsoft's MAI-Image-2.7-Preview takes #2 on image editing arena, closing in on GPT Image 2.5 — ArtificialAnlys · 2026-10-27
- GPT-6.1 Terra scores best-in-class 8% on MazeBench at half of Sol's cost — patience_cave · 2026-10-27