RicursiveAI Releases From-Scratch Dense Ripple-65B, Tops 80% on Terminal Bench 2.1
DZhang50 · x · 2026-11-06
RicursiveAI released Ripple-65B Preview, a 65B dense LLM pretrained entirely from scratch, scoring over 80% on Terminal Bench 2.1 and setting a new US open-weight SOTA for agentic coding, beating significantly larger models including Muse Spark Open-816B, Beam-501B, Inkling-975B, and Nemotron 3 Ultra-550B.
More from coding & agent
- Qwen 3.8 Flash tested on an RTX 3080 Ti: builds an Uber demand simulation end to end — dreamai87 · 2026-11-06
- Clawdock is an open-source Windows tool that turns your Copilot key into a Claude voice assistant — Necessary-Aioli-8402 · 2026-11-06
- LLM agent writes PLC ladder logic offline for under 100K tokens: industrial automation meets coding agents — Many_Somewhere3863 · 2026-11-06
- Ofir Press: No good human-in-the-loop SWE benchmark exists; two futures for agents — OfirPress · 2026-11-06
- Collinear AI runs full-scale agentic RL on Together Post-Training with self-onboarding — togethercompute · 2026-11-06
- YC boosts stacktrace_ai launch: stop monitoring agents, start steering them in real time — ycombinator · 2026-11-06