Together AI open-sources tev1, a decision model finetuned on Qwen3.5 4B with full data recipe
nutlope · x · 2026-09-24
Together AI open-sourced tev1-4B-experimental, a "decision model" finetuned on top of Qwen3.5-4B: give it context, a question, and 2–24 options, and it returns one answer letter. The repo includes the full data recipe, training example, and saved results so you can finetune your own small decision/classification model — or point a coding agent at it and have it do the finetuning for you.
More from coding & agent
- Nyx Terminal launches: a terminal built for managing multi-agent workflows — henrymodis · 2026-09-24
- Waterloo dropout turned OpenAI MTS trains agents to click, type and use any software — CatAstro_Piyush · 2026-09-24
- Dev uses RGBA 4 channels to store 4 depths for a fluid rendering trick — Michael_Moroz_ · 2026-09-24
- How Context Compaction Works: Summarization That Shrinks Long Conversations Without Losing Key Points — blaizedsouza · 2026-09-24
- Builder's Guide: How to Build an Agentic Harness with Jev, Code and Diagrams Included — blaizedsouza · 2026-09-24
- Salesforce Paper: Only 35.8% of the Cleanest Public RL Environments Pass Audit — dair_ai · 2026-09-24