Baseten Loops goes GA with serverless always-warm infra for agent RL training
baseten · x · 2026-11-07
Baseten announced that Loops, its SDK for SFT and RL training of agents, is now GA with a serverless offering. Datadog runs multi-turn agentic RL on it, LangChain post-trains on agent traces for LangSmith Engine, and AmbienceAI tunes models for clinical language and transcription.
Serverless Loops includes:
- Always-warm infrastructure that eliminates cold starts
- Pay-per-token pricing for both training and sampling
- One-click deployment from trained checkpoints to production inference
- Model-specific numerical validation focused on reducing train/sampling mismatch
Teams bring their own data, reward functions, and training logic; Baseten handles infrastructure, and users always own the weights.
Related event: Baseten Loops Goes GA With Serverless for Agent RL Training(3 posts)→
More from coding & agent
- Pair programming with Codex listening in: design debates yield simpler agent code — angadsg · 2026-11-07
- GlobeBench: best of 20 LLMs hits only 49.6% as RL environment simulators — mariyaivasileva · 2026-11-07
- X launches new Livestreaming API routes, old ones EOL Jan 1, 2027 — Kyrannio · 2026-11-07
- Agents are like Memento: your system prompt is their tattoo, most context never reaches them — brandon_galang · 2026-11-07
- Hacker gets Hermes Agent running on a Nintendo 3DS with voice input via custom homebrew — alexcovo_eth · 2026-11-07
- Developer's bar for AI code: if you can't explain it in review, the PR isn't done — therealdanvega · 2026-11-07