Alibaba Qwen confirmed Qwen3.8-Flash is now available in the open-source coding agent OpenCode Go. The model is a 125B-parameter multimodal MoE with about 6B active per token and up to 1M context. That puts a cheap, long-context, multimodal coding model on a day-one path inside an existing agent harness instead of a custom inference stack.
Key Takeaways
- ✓Qwen3.8-Flash is selectable in OpenCode Go as a 125B/6B multimodal MoE with 1M context.
- ✓The open-source coding agent can call it without a custom inference stack.
- ✓Target jobs include coding assistants, agent workflows, and long-context codebase analysis.
⚡
TL;DR VerdictKey Decision Metrics at a Glance
🧭
Turn your technical choice into a development budget
Compare 40 dev plans & simulate token costs vs $20/mo subscriptions
Action HubReady to adopt this in production?
Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.
Up NextMoonshot launches Kimi K3.1 Pro: 2.8T-total / 104B-active open-weight MoE with 1M context, 65.7% on Terminal-Bench 4.0, API at $2/$10, weights by Nov 4
Read Now→
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.