Models
Models
Models shape how your agent reasons, writes, and decides what to do next. Choose a model based on the work you expect the agent to handle: everyday triage, deep research, long-context planning, code changes, or fast lightweight responses.
The model picker includes a short list of featured models and a More models menu for additional options. Availability can vary by workspace; models marked unavailable in the picker cannot be selected.
The Model ID is what you pass as model in an agent spec through the Universal Harness API, or as default_model when creating an agent directly.
Featured models
| Model | Model ID | Provider |
|---|---|---|
| GPT-6.1 Sol | gpt-6.1-sol | OpenAI |
| GPT-6 Astra | gpt-6-astra | OpenAI |
| GPT-6 Sol | gpt-6-sol | OpenAI |
| GPT-6 Luna | gpt-6-luna | OpenAI |
| GPT-5.6 Sol | gpt-5.6-sol | OpenAI |
| GPT-5.6 Terra | gpt-5.6-terra | OpenAI |
| GPT-5.6 Luna | gpt-5.6-luna | OpenAI |
| Claude Opus 5.5 | claude-opus-5-5 | Anthropic |
| Claude Opus 4.8 | claude-opus-4-8 | Anthropic |
| Claude Sonnet 5.5 | claude-sonnet-5-5 | Anthropic |
| Claude Sonnet 5 | claude-sonnet-5 | Anthropic |
| Claude Haiku 5.5 | claude-haiku-5-5 | Anthropic |
| Gemini 3.8 Flash | gemini-3.8-flash | |
| Grok 4.7 | grok-4.7 | xAI |
| Muse Spark 1.3 | meta/muse-spark-1.3 | Meta |
GPT-6.1 Sol and GPT-6 Astra are available for agents using the Kafka, Codex, and OpenCode harnesses. GPT-5.6 Sol remains the default. Both use Standard execution; Fast is not available. They are not available in Kafka Legacy, Qwen, or harnesses limited to other providers' models.
More models
| Model | Model ID | Provider |
|---|---|---|
| GPT-5.5 | gpt-5.5 | OpenAI |
| GPT-5.4 | gpt-5.4 | OpenAI |
| GPT-5.2 | gpt-5.2 | OpenAI |
| GPT-5 | gpt-5 | OpenAI |
| GPT-5 Mini | gpt-5-mini | OpenAI |
| GPT-5 Nano | gpt-5-nano | OpenAI |
| Claude Sonnet 4.6 | claude-sonnet-4-6 | Anthropic |
| Claude Opus 4.7 | claude-opus-4-7 | Anthropic |
| Claude Opus 4.6 | claude-opus-4-6 | Anthropic |
| Claude Opus 4.5 | claude-opus-4-5-20251101 | Anthropic |
| Claude Sonnet 4.5 | claude-sonnet-4-5-20250929 | Anthropic |
| Claude Sonnet 4 | claude-sonnet-4-20250514 | Anthropic |
| Claude Haiku 4.5 | claude-haiku-4-5-20251001 | Anthropic |
| Gemini 3.5 Flash | gemini-3.5-flash | |
| Gemini 3.0 Flash | gemini-3-flash-preview | |
| Gemini 2.5 Pro | gemini-2.5-pro | |
| Gemini 2.5 Flash | gemini-2.5-flash | |
| Grok 4.5 | grok-4.5 | xAI |
| Grok 4 | grok-4 | xAI |
| Grok 4 Fast | grok-4-fast | xAI |
| Muse Spark 1.2 | meta/muse-spark-1.2 | Meta |
| GLM 5.2 | zai-org/GLM-5.2 | Baseten |
| GLM 5.3 | zai-org/GLM-5.3 | Baseten |
| GLM 5.3 Flash | zai-org/GLM-5.3-Flash | Baseten |
| DeepSeek V4 | deepseek-ai/DeepSeek-V4-Pro | Baseten |
| Kimi K2.7 Code | moonshotai/Kimi-K2.7-Code | Baseten |
Baseten-hosted models route through Baseten's OpenAI-compatible endpoint and reason on their own — they don't take an effort setting.
GPT-4o and GPT-4o Mini are retired. Existing saved selections and API or CLI requests that specify a retired model ID run and are billed as GPT-5.6 Luna. Luna has its own behavior and pricing. Use gpt-5.6-luna when replacing either retired model ID in a configuration.
gpt-5.6-luna, gpt-5, and gpt-5-mini do not support caller-supplied sampling settings. Requests that include temperature or top_p for these models have both values omitted before the request is sent, and the model's own defaults apply. Other models, including newer GPT-5 versions, keep caller sampling values.