來源Qwen (GitHub Releases: qwen-code)•較早收集於 4m
Qwen Code v0.10.5 夜間版本發布
#nightly-build#github-release#changelogqwen-codeqwenqwen-code
💡Qwen Code nightly: early fixes post-v0.10.6 for faster coding AI iteration
⚡ 30 秒速覽
有什麼變化
發布夜間版本 v0.10.5-nightly.20260225.229c4431
為什麼重要
此夜間更新為 Qwen Code 使用者提供最新程式碼改進的早期存取,加速編碼任務的迭代。
下一步行動
Review the GitHub changelog at qwen-code releases for v0.10.5-nightly and test in your coding workflows.
誰應關注:Developers & AI Engineers
關鍵要點
- •發布夜間版本 v0.10.5-nightly.20260225.229c4431
- •變更日誌比較 v0.10.6 至此版本
- •可在 Qwen qwen-code 儲存庫的 GitHub Releases 取得
🧠 深度解析
背景與延伸:來自公開資料,非原文內容。引用 8 個來源。
🔑 增強重點摘要
- •Qwen Code is an open-source AI agent designed to live in the terminal, optimized specifically for Qwen-Coder models[7][8].
- •Free usage tier available via Qwen OAuth, providing 2,000 requests per day and 60 requests per minute rate limit[8].
- •Available as an Arch Linux package (qwen-code 0.10.5-1) in the Extra repository, with 9.2 MB package size[7].
🛠️ 技術深入
- •Supports session management with configurable token limits (default example: 32,000 tokens) via .qwen/settings.json[8].
- •Includes commands like /compress for history compression, /clear to reset conversation, and /stats for token usage[8].
- •Features git automation (e.g., analyze commits, create changelogs), file operations (e.g., image conversion, rename files), debugging (e.g., performance bottlenecks), and security audits[8].
- •Benchmarked on Terminal-Bench: 37.5% accuracy with Qwen3-Coder-480A35 model, 31.3% with Qwen3-Coder-30BA3B[8].
🔮 前景展望基於引用來源的 AI 分析
Nightly releases enable rapid iteration on features like OAuth and session management
Arch Linux packaging accelerates adoption among Linux users
Official inclusion in Extra repository as of 2026-02-21 provides easy installation via pacman for x86_64 systems[7].
⏳ 時間線
2026-02
v0.10.5-nightly.20260224.a13d88ac released with version bump and release workflow fixes
2026-02
v0.10.4-nightly.20260220.c6a723ef released with source downloads
2026-02
Arch Linux package qwen-code 0.10.5-1 built and published
2026-02
v0.10.1-nightly.20260212.fb9f3fb4 released with authentication docs update
📎 來源 (8)
Factual claims are grounded in the sources below. Forward-looking analysis is AI-generated interpretation.
📰
AI 週報
閱讀本週精選 AI 大事摘要 →
👉相關動態
AI 策展新聞聚合。所有內容版權歸原始發布者所有。
原始來源: Qwen (GitHub Releases: qwen-code) ↗
每週電子報
每週一封,可隨時退訂。