Collinear AI runs full-scale agentic RL on Together Post-Training with self-onboarding
togethercompute · x · 2026-11-06
Together AI highlights customer Collinear AI, which has built on its Post-Training platform since the April beta. The latest release's self-onboarding let Collinear quickly launch full-scale agentic RL runs, with the team calling the service "exceptionally intuitive, flexible, and powerful for scaling sophisticated agentic workloads."
Together Post-Training's pitch: users own the training loop via a simple API — data, prompts, environments, rewards, verifiers, GRPO/PPO/custom losses, multi-LoRA or full-weight, and deployment target — while Together handles GPU capacity, training-rollout consistency, parallelism, and per-model-tuned rollout engines. One RL step is sample() → reward → forwardbackward() → optimstep() → weightssync(), with parallel experiment sessions and Weights & Biases tracking.
More from coding & agent
- Executor v2 launches: Zapier/Retool-like automation fully drivable by code and AI agents — threepointone · 2026-11-06
- Y Combinator-backed Reflex raises $14M Series A, its Python framework hits 29k GitHub stars — ycombinator · 2026-11-06
- One diagram maps all the agent protocols: MCP, A2A, UCP, AP2, A2UI, AG-UI — rseroter · 2026-11-06
- Sutra: open-source terminal coding agent, lowest cost per solved task — cneuralnetwork · 2026-11-06
- From Greasemonkey to now: AI lets you mod any app just by describing the change — msfeldstein · 2026-11-06
- AI agent handling an office move spams movers with hundreds of questions until they refuse to reply — flngr · 2026-11-06