Available Models
| Model | Config Name(s) | Description |
|---|---|---|
| Claude Opus 4.6 | opus, claude-opus-4.6 | Most capable — best for complex, multi-step tasks (default) |
| Claude Opus 4.5 | claude-opus-4.5 | Previous generation Opus |
| Claude Sonnet 4.5 | sonnet, claude-sonnet-4.5 | Fast and capable — good balance of speed and quality |
| Claude Sonnet 4 | claude-sonnet-4 | Previous generation Sonnet |
| SWE 1.5 | swe, swe-1.5 | Fastest model, at up to 950 tokens/sec |
| SWE 1.5 Free | swe-1.5-free | Free tier of SWE 1.5 |
| Codex 5.3 | codex, codex-5.3 | OpenAI Codex model optimized for code |
| Gemini 3 Pro | gemini, gemini-3-pro | Google’s pro-tier model |
| Gemini 3 Flash | gemini-3-flash | Google’s fast model |
Short names like
opus, sonnet, swe, codex, and gemini always resolve to the latest version in that model family.Reasoning / Thinking Levels
Some models support configurable reasoning levels, which control how much compute the model spends “thinking” before responding. You can set the thinking level with/thinking during a session.
Codex 5.3
| Level | Description |
|---|---|
| Low | Low reasoning |
| Medium | Medium reasoning (default) |
| High | High reasoning |
| X-High | Maximum reasoning |
Claude Models
Claude Opus 4.6, Opus 4.5, and Sonnet 4.5 support thinking at Off and High levels.Setting the Model
- Command flag
- Slash command
- Config file
Model Selection Tips
Complex refactoring
Use
opus for multi-file refactors, architecture changes, and tasks requiring deep reasoning.Quick edits
Use
swe for straightforward edits, bug fixes, and questions — it’s the fastest model.Cost-sensitive
Use
swe-1.5-free when budget matters.