Two-Person Team Uses AI to Generate 20 Inference Engines in Two Weeks, Hitting 1300 tok/s
A two-person team, INT21, used AI to auto-generate 20 inference engines in two weeks across 15 models, calling it an 'AlphaGo moment.' Their engine runs Xiaomi's 1T-parameter model at 1308 tokens/s, prompting claims that hand-written AI software will soon be history.
2026-09-29 ~ 2026-09-30 · 3 related posts
- INT21 claims 20 AI-generated inference engines in 2 weeks, MiMo hits 1,308 tok/s — bingxu_ · 2026-09-29
- AI-generated inference engine hits 1300 tok/s on Xiaomi's 1T model, beating SGLang and vLLM — bingxu_ · 2026-09-30
- Homegrown inference engine runs Xiaomi's 1T-parameter model at 1300 tok/s — bookwormengr · 2026-09-30