Updated November 18, 2026
Uncensored models in OpenCode.
OpenCode takes any OpenAI compatible provider. One config block adds every uncensored model on OpenUncensored.
Short answer
Add OpenUncensored as an OpenAI compatible provider inopencode.json (in the project, or ~/.config/opencode/opencode.json for every project): base URL https://api.openuncensored.com/v1, your key in OPENUNCENSORED_API_KEY, and the model ids you want. Then run opencode run -m openuncensored/deepseek-v4.1-flash-1017-uncensored. One key reaches 62 uncensored, abliterated and heretic models.Config
Add the provider.
Buy a prepaid key on the credits page with no account (the key is the account), or point your agent at agents.md and let it buy one.
{
"$schema": "https://opencode.ai/config.json",
"providers": {
"openuncensored": {
"name": "OpenUncensored",
"env": ["OPENUNCENSORED_API_KEY"],
"package": "@opencode/ai/providers/openai-compatible",
"settings": { "baseURL": "https://api.openuncensored.com/v1" },
"models": {
"deepseek-v4.1-flash-1017-uncensored": { "name": "DeepSeek V4.1 Flash 1017 Uncensored" },
"glm-5.3-flash-uncensored": { "name": "GLM 5.3 Flash Uncensored" },
"qwen3.8-27b-uncensored-fast": { "name": "Qwen 3.8 27B Uncensored Fast by Heabsy" }
}
}
}
}export OPENUNCENSORED_API_KEY=ou-...
opencode run -m openuncensored/deepseek-v4.1-flash-1017-uncensored "Explain this repository"Add any id from the models page to the models block the same way, or list them all with curl -s https://api.openuncensored.com/v1/models | jq -r '.data[].id'.
Check
Test the key first.
curl https://api.openuncensored.com/v1/chat/completions \
-H "Authorization: Bearer $OPENUNCENSORED_API_KEY" -H "Content-Type: application/json" \
-d '{"model": "deepseek-v4.1-flash-1017-uncensored", "messages": [{"role": "user", "content": "Hello"}]}'If this works and OpenCode does not, the problem is the config or the environment OpenCode runs in.
Gotchas
Things that trip people up.
- The background daemon does not see your shell.
opencode servestarted as a service does not read variables you export later. Put the key where the daemon reads it, or test withopencode run --standalone. - Model names are prefixed with the provider key:
openuncensored/<model id>. - Older versions use the
providerblock with@ai-sdk/openai-compatible(the OpenCode 1.x tab). Base URL and ids are the same.
Which model
Models that finish agent tasks.
| Model | Model id | Agent tasks | Speed | $ / 1M in · out |
|---|---|---|---|---|
| GLM 5.3 Flash Uncensored | glm-5.3-flash-uncensored | 3/5 | 178 tok/s | $0.12 · $0.50 |
| DeepSeek V4.1 Flash 1017 Uncensored | deepseek-v4.1-flash-1017-uncensored | 2/5 | 239 tok/s | $0.33 · $1.32 |
| MiMo V2.6 Flash Uncensored | mimo-v2.6-flash-uncensored | 2/5 | 118 tok/s | $0.46 · $1.50 |
| DeepSeek V4.1 Flash 0910 Uncensored | deepseek-v4.1-flash-0910-uncensored | 2/5 | 209 tok/s | $0.33 · $1.32 |
| GLM 5.3 Uncensored | glm-5.3-uncensored | 2/5 | 117 tok/s | $0.39 · $1.32 |
FAQ
Questions people ask.
How do I add an uncensored model provider to OpenCode?
Add OpenUncensored as an OpenAI compatible provider in opencode.json (in the project, or ~/.config/opencode/opencode.json for every project): base URL https://api.openuncensored.com/v1, your key in OPENUNCENSORED_API_KEY, and the model ids you want. Then run opencode run -m openuncensored/deepseek-v4.1-flash-1017-uncensored. One key reaches 62 uncensored, abliterated and heretic models.
Which uncensored model is best in OpenCode?
GLM 5.3 Flash Uncensored: the most agent tasks finished among ranked models fast enough for agent loops.
Does OpenCode's tool calling work with uncensored models?
Yes, through the standard OpenAI tools format. How reliably a model uses tools differs; the agent tasks column on this page measures it.
Do I need an account?
No. Buy a prepaid key on the credits page and use it as the API key. The key is the account, and it cannot spend more than you loaded.
Updated November 18, 2026. Scores measured November 8, 2026. Markdown version: /opencode-uncensored-provider.md