RWKV-8 RPU preview hits 12,500+ tokens/s real-time inference
kastnerkyle · x · 2026-11-17
RWKV team shared a preview of RWKV-8 RPU running real-time intelligence at 12,500+ tokens/s. The model is still early, and the team aims to run leading OSS flash models at similar speeds by 2027, with more demos of increasingly larger models coming soon.
More from Infra
- Lambda completes world's first large-scale CPO deployment linking 10,000+ Blackwell Ultra GPUs — TheZachMueller · 2026-11-17
- DeepSeek Flash on Mac Studio M5 Ultra (512GB): 54 t/s generation with 128K context — vhthc · 2026-11-17
- Peter Diamandis: The Country Building Gigawatts Fastest Will Set the Price of Intelligence — PeterDiamandis · 2026-11-17
- Gemini Managed Agents Adds Self-Hosted Sandboxes: Google Runs the Loop, You Run the Tools — _philschmid · 2026-11-17
- Your AI budget is too generous and too restrictive at once: routing saves, flat caps kill rewrites — ruslansv · 2026-11-16
- Edge inference with llm-d: prefill-decode disaggregation on Qualcomm Cloud AI 100 Ultra — HowDevelop · 2026-11-16