Kimi K3.1 Pro on Cerebras hits 1375 TPS after 594 agent-run experiments
hyperparticle · x · 2026-11-23
Developers venkatGoku and areseyefu spent a weekend pushing inference speed limits with Kimi K3.1 Pro on Cerebras hardware, using OpenClaw on the Santa's Speedy Shopper challenge. Their best verified run hit 1375 mean TPS (peak 1795 TPS per model call) while finding 5 real products from live shopping sites. Behind the result: 143 research waves and 594+ experiments coordinated via the Rekursiv AI stack.
More from coding & agent
- User asks Anthropic to publish a full human-agent trace of solving a Millennium Prize problem — _arohan_ · 2026-11-23
- MemTensor launches MemT-1.0-Pro, a 554B memory-native model built with NVIDIA — _akhaliq · 2026-11-23
- Team coordinates AI research agents with a shared research graph — hyperparticle · 2026-11-23
- Formalizing CFSG at 3x RH scale: Lean is now the bottleneck for AI math — Vjeux · 2026-11-23
- Quote-gate + string checker: agent memory got reliable once the author stopped trusting it — capngreenjeans · 2026-11-23
- 'Permission is the enemy of autonomy': devs slam Codex approval loops over worktree cleanup — Baconbrix · 2026-11-23