Meta's Muse Spark Open goes live on Baseten Loops SDK for reward-based post-training
baseten · x · 2026-10-26
Baseten announced that Meta's Muse Spark Open is live on its Loops SDK for post-training: bring your own data and reward functions, and Loops runs the training and sampling infrastructure, with checkpoints deployable straight to production inference endpoints. The quickstart docs walk through creating a LoRA training session (trainer + sampler), GPU-time billing, and shutdown, using Qwen3.5-2B as the example base model.
More from Infra
- Sam Altman: robots building robots and data centers building data centers form another AI flywheel — haider1 · 2026-10-26
- Middle schooler hand-writes Metal RMSNorm kernel for own inference engine on Apple Silicon — dosco · 2026-10-26
- Default-to-zero: a model-name mismatch silently wiped all LLM costs to $0 — Atm1n9 · 2026-10-26
- Dev: I regret buying software licenses, should have bought more compute — cephaloform · 2026-10-26
- Musk: 200+ neoclouds face 'guppy-level death' as debt markets shut for AI-only firms — Yuchenj_UW · 2026-10-26
- Tailscale VP of Engineering responds to SSH vuln concerns: no need to disable it yet — basedjensen · 2026-10-26