GLM 5.3 Flash Cyberin opencode
Yes. GLM 5.3 Flash Cyber runs in opencode through this endpoint as glm-5.3-flash-cyber. Security-tuned, not uncensored. GLM 5.3 Flash tuned for security work: code review, exploit analysis and incident response. A million tokens of context, tool calling and vision. Security-tuned, not an uncensored model.
- Model id
glm-5.3-flash-cyber- Build
- Security-tuned
- Context
- 1M tokens
- Tools / reasoning / images
- Yes / Yes / Yes
- Per million tokens
- On the models page
Where it fits
GLM 5.3 Flash tuned for security work: code review, exploit analysis and incident response. Pick it when the session is an audit or a triage and you want answers written the way a security engineer would write them.
What to watch for
This is the one model on the line that is not an uncensored build, and we say so wherever it is listed. If a task gets refused here, switch to GLM 5.3 Flash Uncensored with /models and carry on in the same session.
The config
{
"$schema": "https://opencode.ai/config.json",
"provider": {
"uncensored-opencode": {
"npm": "@ai-sdk/openai-compatible",
"name": "Uncensored Opencode",
"options": {
"baseURL": "https://uncensoredopencode.com/v1",
"apiKey": "{env:UNCENSORED_OPENCODE_API_KEY}"
},
"models": {
"glm-5.3-flash-cyber": {
"name": "GLM 5.3 Flash Cyber",
"tool_call": true,
"reasoning": true,
"attachment": true,
"limit": { "context": 1048576, "output": 32000 }
}
}
}
},
"model": "uncensored-opencode/glm-5.3-flash-cyber"
}Your key goes in UNCENSORED_OPENCODE_API_KEY. The opencode page has the setup steps and the block with every model in it.
A first task for it
opencode run --model uncensored-opencode/glm-5.3-flash-cyber "Audit this codebase for injection and auth flaws and rank what you find by severity"Next to it on the line: GLM 5.3 Flash, Kimi K3.1 Pro, GLM 5.3, DeepSeek V4.1 Flash 1017, MiMo V2.6 Flash, Qwen3.8 27B.