opencode is an open-source, terminal-based AI coding agent. It supports OpenAI-compatible endpoints and can be pointed directly at the L&S AI Inference API.
More information is available here: API Access
export OPENAI_API_KEY="your-lsit-api-key"
export OPENAI_BASE_URL="https://api.ai.college.ucsb.edu/v1"
opencode
Add these to your shell profile (~/.bashrc, ~/.zshrc, etc.) to make them persistent.
| Model ID | Display name | Best for | Context |
|---|---|---|---|
Qwen3-Coder-Next |
Qwen3 Coder Next | Code generation, editing, agentic coding | 128k |
gpt-oss-120b |
GPT-OSS 120B | Complex reasoning, planning, large context synthesis | 128k |
gemma-4-31b |
Gemma 4 31B | General chat, light tasks, subagent use | 128k |
All models support tool/function calling. Responses are buffered (streaming disabled) — this is expected behavior.
Reference models in config as ai/<model-id>, e.g. ai/Qwen3-Coder-Next.
opencode.json — it may cache provider config between sessions.gpt-oss-120b and keeping file-level edits on Qwen3-Coder-Next.