nsheaps/glm-models
plugins/zai-glm/skills/glm-models/SKILL.md
Use this skill when the user asks about GLM models, GLM-5, GLM-4.7, GLM-4.6, GLM-4.5, GLM-4V, ChatGLM, CogView, CogVideoX, z.ai model capabilities, model selection for different tasks, or comparing GLM models.
$ install --global
skillsauthnpx skillsauth add nsheaps/ai-mktpl glm-modelsInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Apr 22, 2026, 4:19 PM36.5s1 file scanned
SKILL.md
- name:
- glm-models
- description:
- >
GLM Model Family
The GLM (General Language Model) family is developed by z.ai (formerly Zhipu AI / 智谱AI). These models support text generation, vision, code, embeddings, image generation, and video generation. All recent models are open-weight under MIT license.
- Docs: https://docs.z.ai/
- API Base URL:
https://api.z.ai/api/paas/v4/ - Pricing: https://docs.z.ai/guides/overview/pricing
Flagship Text Models
| Model | Architecture | Context | Key Features |
| --------------- | ---------------------- | ------------------ | ------------------------------------------------------------- |
| glm-5 | ~745B MoE (44B active) | 200K in / 128K out | Agentic engineering, tool streaming, long-horizon tasks, MIT |
| glm-5-turbo | Same, optimized | 200K in / 128K out | Improved stability for long-chain agent tasks |
| glm-4.7 | ~400B MoE | 200K in / 128K out | Coding-focused, Preserved Thinking, Turn-level Thinking, MIT |
| glm-4.7-flash | Lightweight | Reduced | Free tier, lighter capability |
| glm-4.6 | 355B total | 200K | Strong code benchmarks, agent frameworks, MIT |
| glm-4.5 | 355B / 32B active | 128K | Hybrid reasoning (thinking/non-thinking modes), deep thinking |
| glm-4.5-x | Premium tier | 128K | Higher capability, premium pricing |
| glm-4.5-air | 106B / 12B active | 128K | Compact variant of GLM-4.5 |
| glm-4.5-flash | Lightweight | 128K | Free tier |
Thinking Mode
GLM-4.5+ models support hybrid reasoning — toggle between deep thinking and instant response:
{
"model": "glm-4.7",
"messages": [{ "role": "user", "content": "Solve this step by step" }],
"thinking": { "type": "enabled" }
}
- Preserved Thinking (GLM-4.7): Retains thinking blocks across multi-turn conversations
- Turn-level Thinking (GLM-4.7): Per-turn control — disable for lightweight requests, enable for complex tasks
- Tool Streaming (GLM-5): Stream output during tool calling (
tool_stream: true)
Vision / Multimodal Models
| Model | Parameters | Context | Description |
| ---------------- | ----------------- | ------- | -------------------------------------- |
| glm-4.6v | 106B / 12B active | 128K | Vision understanding, function calling |
| glm-4.6v-flash | 9B | — | Free, open weights, commercial license |
| glm-4.5v | 106B VLM | — | Vision-language model |
Vision API Example
curl "https://api.z.ai/api/paas/v4/chat/completions" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "glm-4.6v",
"messages": [{
"role": "user",
"content": [
{"type": "text", "text": "What is in this image?"},
{"type": "image_url", "image_url": {"url": "https://example.com/image.jpg"}}
]
}]
}'
Specialized Models
| Model | Category | Description |
| ----------------- | ---------------- | -------------------------- |
| glm-image | Image generation | Text-to-image (Jan 2026) |
| glm-ocr | OCR | Document and image OCR |
| cogview-3-plus | Image gen | High-quality text-to-image |
| cogvideox | Video gen | Text-to-video generation |
| cogvideox-flash | Video gen | Fast video generation |
Embedding Models
| Model | Dimensions | Description |
| ------------- | ---------- | ------------------------------- |
| embedding-3 | 2048 | General-purpose text embeddings |
| embedding-2 | 1024 | Previous generation embeddings |
curl "https://api.z.ai/api/paas/v4/embeddings" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "embedding-3",
"input": "What is machine learning?"
}'
Model Selection Guide
| Use Case | Recommended Model | Why |
| ------------------- | ----------------- | --------------------------------------- |
| Agentic tasks | glm-5 | Tool streaming, long-horizon planning |
| Coding | glm-4.7 | Coding-focused, Preserved Thinking |
| Complex reasoning | glm-4.5 | Hybrid reasoning with deep thinking |
| General chat | glm-4.5-flash | Free, good quality |
| High throughput | glm-4.5-air | Compact, fast inference |
| Image understanding | glm-4.6v | Best vision model with function calling |
| Embeddings/search | embedding-3 | Latest generation |
| Image creation | glm-image | Latest generation (Jan 2026) |
| Budget-conscious | glm-4.5-flash | Free tier available |
Claude Code Model Mapping
When using z.ai's Anthropic-compatible endpoint with Claude Code, map models to slots:
| Claude Code Slot | Recommended GLM Model | Rationale |
| ---------------- | --------------------- | ---------------------------- |
| Opus | glm-5 | Most capable, agentic |
| Sonnet | glm-4.7 | Strong coding, balanced cost |
| Haiku | glm-4.5-air | Fast, cost-effective |
Pricing (per 1M tokens, USD)
| Model | Input | Output |
| ---------------- | ------ | ------ |
| glm-5 | ~$1.00 | ~$3.20 |
| glm-4.7 | $0.60 | $2.20 |
| glm-4.7-flash | Free | Free |
| glm-4.5 | ~$0.20 | ~$1.10 |
| glm-4.5-x | — | $8.90 |
| glm-4.5-flash | Free | Free |
| glm-4.6v | ~$0.14 | ~$0.41 |
| glm-4.6v-flash | Free | Free |
Prices approximate; see docs.z.ai/guides/overview/pricing for current rates. Batch API available at 50% cost.
Unique Features
- MIT license: GLM-4.5, 4.6, 4.7, and 5 are all open-weight under MIT
- 200K context: GLM-4.6, 4.7, and 5 support 200K input with up to 128K output
- Hybrid reasoning: Toggle deep thinking on/off per request or per turn
- Tool streaming: GLM-5 streams output during tool calls for real-time agent UX
- Free tiers:
glm-4.5-flash,glm-4.7-flash,glm-4.6v-flashare free - Domestic chip training: GLM-5 trained on Huawei Ascend chips, GLM-4.6 on Cambricon — zero NVIDIA dependency
- Bilingual strength: Particularly strong in Chinese + English tasks
- Anthropic-compatible API: Native Claude Code integration without proxies
- Native function calling: OpenAI-style tool description format in all recent models
References
- z.ai Developer Docs
- GLM-5 Overview
- GLM-4.7 Docs
- GLM-4.6 Docs
- GLM-4.5 Docs
- Pricing
- z.ai on GitHub
- z.ai on Hugging Face
Related Skills
nsheaps/github-app-session-env
tools
VerifiedTrustedCommunity
Manually reproduce what the github-app plugin's SessionStart hook does to make a GitHub App installation token usable in the current session — materialize the PEM, generate the token, isolate GH_CONFIG_DIR, write the runtime env file, and wire CLAUDE_ENV_FILE so every Bash call sees GH_TOKEN/GITHUB_TOKEN. Use when the hook did not run, the token is missing from the environment, or a shell/teammate needs the token wired up by hand. <example>GH_TOKEN isn't set even though github-app is configured</example> <example>the github-app SessionStart hook didn't run, set up the token manually</example> <example>wire the github app token into CLAUDE_ENV_FILE</example> <example>gh keeps falling back to the wrong account, isolate GH_CONFIG_DIR</example>
3SKILL.mdUpdated Jun 9, 2026
nsheaps/github-app-git-identity
tools
VerifiedTrustedCommunity
Manually configure the GitHub App bot git identity the way the github-app plugin's SessionStart hook does — resolve the app slug and bot user ID, build the <slug>[bot] name and noreply email, set GIT_AUTHOR_*/GIT_COMMITTER_* env vars, and write an isolated GIT_CONFIG_GLOBAL with the gh auth git-credential helper. Use when commits are attributed to the wrong account, "Author identity unknown" appears, or git identity must be set up by hand. <example>my commits are showing up as the handler, not the bot</example> <example>git says Author identity unknown after the github-app hook ran</example> <example>configure the github app bot git identity manually</example> <example>set up the gh credential helper for git push</example>
3SKILL.mdUpdated Jun 9, 2026
nsheaps/spec-management
tools
VerifiedTrustedCommunity
Manages spec files for requirements capture and validation
3SKILL.mdUpdated Jun 7, 2026
nsheaps/plugins/bash-command-rejection/skills/bash-chaining-alternatives
tools
VerifiedTrustedCommunity
# Bash Chaining Alternatives This skill teaches you how to work around the bash command chaining restriction enforced by this plugin. ## Why Chaining is Blocked The `bash-command-rejection` plugin blocks these operators: | Operator | Name | Why Blocked | | -------- | ---------- | ----------------------------------------------------------------------------------- | | `&&` | AND chain | Runs cmd2 only if cmd1 su
3SKILL.mdUpdated Jun 7, 2026