Valen on GitHub: train a Jev-like visual decision model yourself
aigclink · x · 2026-09-28
The GitHub repo for Valen (Liuziyu77/Valen, 390 stars) is now public. Inspired by Jev, it ingests text, images and video and returns probabilities over supplied candidates via a Qwen3.5-0.8B/2B backbone plus a shared decision head — no answer tokens generated. The repo ships model implementation, data processing, SFT and experimental RLCD training, inference and evaluation commands, and supports training on your own data. An online demo (Valen-Preview-0923) shows it reading board images and picking moves against a 27B generation model.
Related event: Open-Source Visual Decision Model Valen Launches with 122ms Decisions(2 posts)→
More from coding & agent
- Polyphonic's Mnemos memory layer is coming to ChatGPT and Claude apps — RileyRalmuto · 2026-09-28
- Obsidian Starter Kit v5 ships: an AI-ready vault built for Claude Code — dSebastien · 2026-09-28
- Keep Claude Code and Codex running with the lid closed using free Amphetamine app — JeremyNguyenPhD · 2026-09-28
- Open-sourced Claude skill finds sales leads for $0.0002 per signal, replacing $167/mo Clay plans — SimplyAnnisa · 2026-09-28
- DeepMind's Veo/Gemini Omni lead rebuilds his site with Antigravity—'it just works' — dumierhan · 2026-09-28
- Always-on agents from Anthropic and OpenAI will be big for engineering—but hard to monetize beyond it — cnakazawa · 2026-09-28