zai-org/GLM-5

GLM-5: From Vibe Coding to Agentic Engineering

Stars: 7.2k
Forks: 950
Open beginner issues: 0
Indexed issues: 32
Avg merge: 1h 33m
Merged PRs (30d): 2

Dominant language: No language data
License: Apache-2.0
Last GitHub push: Sep 1, 2026
Latest indexed: Sep 20, 2026
Contributing guide: No contributing guide
Code of conduct: No code of conduct
Beginner labels: help wanted

Beginner friendly Everything

Open All

32 open issues indexed

Doestn`t see the files Open

Difficulty 3/5 1-2 days Newbie friendliness 38/100

zai-org/GLM-5#156 · Opened Sep 20, 2026
Verify evals on Papers with Code Open

Difficulty 4/5 3-5 days Newbie friendliness 45/100

zai-org/GLM-5#155 · Opened Sep 18, 2026
Enhancement: Add API usage dashboard with token, cache, and cost breakdown Open

enhancement

zai-org/GLM-5#154 · 1 assignee · Opened Sep 15, 2026
在智能体工具使用的时候有问题 Open

zai-org/GLM-5#153 · 3 comments · 1 assignee · Opened Sep 15, 2026
清言App支持跨会话记忆、个性化指令与本地知识库(RAG) Open

enhancement

zai-org/GLM-5#152 · 1 comment · 1 assignee · Opened Sep 13, 2026
[Coding Plan] MCP tools (web_search_prime / web_reader) hit 429 rate limit aggressively — thresholds undocumented, unrelated to quota Open

zai-org/GLM-5#151 · 1 comment · 1 assignee · Opened Sep 12, 2026
GLM API Usage Bundle purchased and confirmed by payment receipt, but account balance shows 0.00 — error 1113 on all requests Open

zai-org/GLM-5#148 · 1 assignee · Opened Sep 11, 2026
GLM-5.3-Flash: reasoning_content padded with repeated "Hmm" fillers Open

zai-org/GLM-5#146 · 2 comments · 1 assignee · Opened Sep 11, 2026
[Bug] glm-5.3-flash: sending `tools` adds a flat ~3s to first token, per request — independent of tool count, tool_choice, and a 100% prompt-cache hit Open

zai-org/GLM-5#144 · 1 comment · 1 assignee · Opened Sep 5, 2026
[Bug] GLM-5.3-Flash spontaneously emits Chinese text mid-response in English agentic contexts (survives context reset) Open

zai-org/GLM-5#142 · 4 comments · 2 reactions · 1 assignee · Opened Sep 4, 2026
Anthropic-compatible API: `message_start` reports zero `usage.input_tokens`, breaking clients that read context usage at stream start Open

zai-org/GLM-5#139 · 1 assignee · Opened Sep 1, 2026
Release GLM-5.3 under a FLOSS license Open

Difficulty 5/5 Over a week Newbie friendliness 25/100

zai-org/GLM-5#135 · 1 comment · Opened Aug 28, 2026
GLM-5.3-Flash quotes a system prompt it was never given on video input, and sometimes answers it instead of the user Open

Difficulty 4/5 3-5 days Newbie friendliness 55/100

zai-org/GLM-5#134 · Opened Aug 28, 2026
[Bug] GLM-5.2: command prefix planned in thinking is dropped from the emitted tool call — unrecoverable repetition loop (10 consecutive failures) Open

zai-org/GLM-5#116 · 4 comments · 1 reaction · 1 assignee · Opened Jul 21, 2026
[GLM-5.2][Safety] Fixed correction-insulin dose suggested when correction factor is unknown Open

zai-org/GLM-5#111 · 2 comments · 1 assignee · Opened Jul 12, 2026
[GLM-5.2] Tool results intermittently rendered as image content blocks instead of text (context-length correlated) Open

zai-org/GLM-5#110 · 1 assignee · Opened Jul 11, 2026
Remote OAuth MCP servers fail with `invalid_token` when defined in config.json (no OAuth provider attached to config-file transport) Open

zai-org/GLM-5#106 · 1 comment · 1 assignee · Opened Jul 8, 2026
what is the difference between the open-source GLM-5.2 and the GLM-5.2 available in Coding Plan? Open

zai-org/GLM-5#97 · 1 comment · 1 assignee · Opened Jun 29, 2026
Proposal: Autonomous Adversarial Pipeline for GLM-5.3 — Failure Taxonomy, Trajectory Farming & IndexShare Stress-Testing Open

zai-org/GLM-5#94 · 1 assignee · Opened Jun 27, 2026
Official Support Channels for GLM API, ZCode, and Related Products / GLM API、ZCode 等产品官方反馈渠道 Open

bug

Difficulty 2/5 1-3 hours Newbie friendliness 38/100

zai-org/GLM-5#93 · 6 comments · Opened Jun 27, 2026
reasoning_content 跨轮对模型不可见，多轮对话中推理链断裂 Open

zai-org/GLM-5#92 · 2 comments · 1 assignee · Opened Jun 27, 2026
[Proposal] Architectural Optimization: Manifold-Constrained Distillation for GLM-5.2 Open

zai-org/GLM-5#88 · 1 assignee · Opened Jun 24, 2026
[Bug] Tool call errors with malformed argument keys: Model tried to call unavailable tool Open

zai-org/GLM-5#84 · 1 assignee · Opened Jun 17, 2026
How to evaluate model on vending bench 2 Open

zai-org/GLM-5#70 · 1 comment · 1 assignee · Opened May 11, 2026
I am unable to use my api Open

zai-org/GLM-5#68 · 10 comments · 1 reaction · 1 assignee · Opened May 6, 2026
[Bug] GLM-5.1 streaming: truncated JSON in SSE chunks causes client-side parse failures Open

zai-org/GLM-5#66 · 5 comments · 1 assignee · Opened Apr 30, 2026
[Bug] web_search_prime MCP returns empty results via Python SDK (mcp package) while works in OpenCode Open

Difficulty 4/5 3-5 days Newbie friendliness 45/100

zai-org/GLM-5#60 · Opened Apr 23, 2026
开启 MTP 预测后，工具调用异常 Open

zai-org/GLM-5#58 · 2 comments · 1 reaction · 1 assignee · Opened Apr 22, 2026
long context respone with blank Open

zai-org/GLM-5#46 · 3 comments · 1 assignee · Opened Apr 3, 2026
[需求] 建议智谱官方 API 支持 OpenAI 兼容的 /responses 接口，以提升 Codex 兼容性 Open

enhancement help wanted

zai-org/GLM-5#39 · 43 comments · 19 reactions · 1 assignee · Opened Mar 30, 2026
How to reproduce the benchmark of GLM-5 on different datasets? Open

zai-org/GLM-5#30 · 1 comment · 1 assignee · Opened Mar 17, 2026
[Bug] Malformed tool-call JSON via NVIDIA NIM (GLM-5) causes parser failures in OpenCode Open

zai-org/GLM-5#15 · 8 comments · 1 assignee · Opened Feb 16, 2026

A short digest of beginner-friendly GitHub issues.