Alibaba Qwen open-sourced Qwen3.8-Flash, a multimodal MoE with 125B parameters plus 51B n-gram embeddings and only 6B active per token. It scores 58.7 on DeepSWE 1.1 and 62.5 on SWE-bench Pro, was trained at 1/9 the cost of Qwen3.7-Plus, and ships Qwen3.8-Flash-Next as an early look at the Qwen4 architecture.

Key Takeaways

  • ✓Scores 58.7 DeepSWE 1.1, 62.5 SWE-bench Pro, and 73.9 CoWorkBench, beating Qwen3.7-Plus especially on coding and office work
  • ✓GDN+QSA hybrid attention, Gated Residual, n-gram embeddings, and Muon; QSA is up to 7.6x faster in 1M-token prefill
  • ✓Weights on Hugging Face and ModelScope with day-0 vLLM/SGLang support; QwenCloud API at $0.16/$0.47 per 1M tokens
⚡

Key Decision Metrics at a Glance

TL;DR Verdict
🏆 Benchmark RankRank
Scores 58.7 DeepSWE 1.1, 62.5 SWE-bench Pro, and 73.9 CoWorkBench, beating Qwen3.7-Plus especially on coding and office work
Full Leaderboard→
💰 Cost & PlansCompare
Qwen3.8-Max · in: $2 / out: $6
Compare Plans↗
🛠️ Ecosystem & IDERules
Cline (VS Code Agent) · Continue · OpenHands (All-Hands AI)
Setup IDE Rules↗
🧭

Turn your technical choice into a development budget

Compare 40 dev plans & simulate token costs vs $20/mo subscriptions

⚔️ Compare Model→💰 Compare plans & limits↗
Action HubReady to adopt this in production?

Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.

⚔️ Launch VS Compare→💳 View Plans↗
Up NextOrnith-1.5-122B Released Under MIT: A Single-Node 122B MoE Reasoning Model Scoring 76.9 on SWE-bench Pro
Read Now→