Qwen Code v0.21.8 Released: Optimizes PR Reviews and Cache Hits
qwen-code-ci-bot · ghdev · 2026-08-09
Tongyi Qianwen's official coding tool, Qwen Code, has released version v0.21.8, introducing several engineering and performance optimizations:
- PR Review Enhancements: Restored real-time autofix support for PRs from forks by bridging review events; added repository context manifests to guide reviews; moved remote matching to a deterministic CLI subcommand, achieving up to 93.3% prompt cache hit rates for /review.
- Cache & Concurrency: Enabled compression cache sharing for OpenAI, Gemini, and other compatible providers to reuse conversation prefixes; allowed ACP agent tool calls to run concurrently within configured limits to improve fan-out operations.
- Security Fix: Patched a vulnerability where explicitly distrusted workspaces incorrectly inherited trust from parent directories.
More from coding & agent
- Constraints you force on models today will become tech debt in months, warns dev — zack_overflow · 2026-10-16
- Liquid AI tutorial: fine-tune your d1-3B decision model in 30 minutes on a free GPU — helloiamleonie · 2026-10-16
- Alibaba open-sources OpenGCUA, a full-stack generalist computer-use agent; +19.6 points on OSWorld-v2 across 8 models — Tongyi-MAI · 2026-10-16
- BotBeat: a social network where all users are AI agents, exposed via 32 MCP tools — Dismal_Piccolo4973 · 2026-10-16
- DKGI paper: agents write their own tools to search knowledge graphs, beating baselines by 6.7 Hit@1 — yuz9yuz · 2026-10-16
- Dev open-sources lgtmaybe: running PR code reviews on a local 27B model via hunk splitting — mattjcoles · 2026-10-16