Together AI adds MiniMax M3.1 Flash post-training with SFT and RL support
togethercompute · x · 2026-10-26
Together AI's Post-Training platform now supports MiniMax M3.1 Flash (256K context). Users can run SFT or RL (GRPO, PPO, custom losses; multi-LoRA or full-weight) with their own data, rewards, and environments while keeping the training loop in their own Python code. Together handles GPU capacity, training-rollout consistency for RL, parallelism, and model-tuned rollout engines, with parallel experiments and W&B-compatible tracking. Kimi-K3 and GLM-5.3 are also available on the platform.
More from Infra
- Anthropic engineer: Claude Code cache hit rates are now high, savings live — teortaxesTex · 2026-10-26
- Deco decouples 'thinking' from 'spelling' in generative recommenders for 3.2-8.6x throughput — _reachsumit · 2026-10-26
- Redditor with $4,000-$5,000 budget seeks local AI video generation hardware advice — dewsphere · 2026-10-26
- ThousandML Open-Sources CQ 4-bit QLoRA: 17.4% Lower Val Loss Than Unsloth NF4 — textclf · 2026-10-26
- Why OpenAI and Anthropic Cut LLM Prices by Up to 80% in Months — Azer_pubes33 · 2026-10-26
- GPU rental rates split by chip generation: B300 +8.6% in a week, A100 only +2% — AccBalanced · 2026-10-26