MiniMax-M3.1-Flash hits ~100 tok/s in speed tests, trailing DeepSeek v4.1-flash
TheZachMueller · x · 2026-10-19
A user benchmarked the new MiniMax-M3.1-Flash via its hosted API: decode speeds ranged from 68.8 to 105.7 tok/s across runs, averaging close to 100 tok/s with no provider errors.
The verdict: slightly slower than DeepSeek-v4.1-flash, but the author notes that 100 tok/s is now effectively the minimum bar for any "flash"-class model. Screenshot shows think-token counts between 1,444 and 1,962 per run.
More from Models
- Users want open-format export from Claude Docs as charts only save as images — BlackHC · 2026-10-20
- EMNLP study: LLM safety interventions drop up to 4.5x when distress is wrapped in delusion — MaartenSap · 2026-10-20
- Cybersecurity-focused VLM OrcaCyber-Nano-1.0 trends on Hugging Face — orcarouter · 2026-10-20
- 0.8B local model with 2,000 labels beats Jev's frontier API on banking data — kalyan_kpl · 2026-10-20
- New TTS benchmark tests 26 models across 30,564 requests: fastest at 46ms — mattturck · 2026-10-20
- Agents handle unreleased language Jai flawlessly, even compile-time execution — lucasmeijer · 2026-10-20