DwarfStar engine hits 170 t/s on Station with MXFP4 optimization
antirez · x · 2026-08-17
Redis creator antirez shared performance benchmarks using the DwarfStar inference engine on Station. By optimizing for roughly 48 hours with MXFP4 and the Flash model, he achieved 170 tokens/sec generation and 22k tokens/sec prefill speeds. He also noted testing PRO and Q4 GLM 5.2 to evaluate how hardware-aware inference engines maintain performance when models exceed VRAM capacity.
Related event: antirez Tunes DwarfStar to 170 t/s Generation on Station(2 posts)→
More from Infra
- Three CoreWeave co-founders have sold $3.3B in shares since IPO amid credit worries — SumitGup · 2026-10-16
- Cherry pi on a Raspberry Pi: the tiny-model rumor developers are watching — sull · 2026-10-16
- CoreWeave's debt costs falling: DDTL spreads tightened from S+9.62% to S+2.25% — BenBajarin · 2026-10-16
- SoftBank lines up US IPOs for SB Energy and AI robotics spinoff Roze, valuations top $50B — nordicinst · 2026-10-16
- TRL reproduces Modal's mixed-precision async RL: plain GRPO collapses, SC+MIS holds 0.72 — QGallouedec · 2026-10-16
- Hugging Face scales Indic pretraining with 2.8T tokens and an open data pipeline — Hugging Face Blog · 2026-10-16