Skip to main content

​
GLMT Controls

GLMT (GLM with Thinking) enables thinking/reasoning mode for GLM models, providing visible reasoning steps before answers.

​
Usage

ccs glmt "complex problem to solve"

​
How It Works

GLMT runs a local proxy that:
  1. Intercepts requests to Claude CLI
  2. Routes to GLM API with thinking enabled
  3. Transforms response to show thinking blocks

​
Configuration

GLMT is automatically configured when you set up a GLM profile. The proxy handles the thinking block transformation.

​
Troubleshooting

​
Thinking Blocks Not Visible

  1. Enable debug logging: 
    export CCS_DEBUG_LOG=1
ccs glmt --verbose "test"
  2. Check raw response: 
    cat ~/.ccs/logs/*response-openai.json | jq '.choices[0].message.reasoning_content'
  3. Scenarios:
    • reasoning_content present: Transformation issue
    • reasoning_content absent: API issue

​
Proxy Startup Timeout

Error: “Proxy startup timeout (5s)” Solutions: 
# Try non-thinking mode
ccs glm "your prompt"

# Check Node.js
node --version  # Requires ≥14

# Check port
netstat -an | grep 127.0.0.1

# Verbose details
ccs glmt --verbose "test"

​
Duplicate Lines

Symptom: Two spinner lines during thinking Cause: Terminal rendering (not CCS issue) Solutions:
  • Use native terminal (iTerm2, GNOME Terminal, Windows Terminal)
  • Exit tmux/screen before running