Choosing a model and reasoning effort
Enable models for your workspace, pick a model and effort in the composer, and understand which credential each provider needs.
This page shows you where models are turned on, how to pick a model and reasoning effort for a session, and what each provider needs before a session can use it.
Enable models in Settings › Models
Settings › Models lists every model in the catalog, grouped by provider, under the heading Enabled Models. Toggle a model on or off, or use Enable all and Disable all on a group. Changes save automatically and apply to the model selector across the web app and the Slack bot. You cannot disable the last enabled model.
| Group | Enabled by default |
|---|---|
| Anthropic | Yes |
| OpenAI | Yes |
| OpenCode Zen | No |
| OpenCode Go | No |
| xAI / SuperGrok | No |
| Z.AI Coding Plan | No |
| DeepSeek | No |
The deployment default model is Claude Sonnet 4.6 (anthropic/claude-sonnet-4-6). It is used when nothing else (a session, an integration setting, or an automation) picks a model.
Pick a model and effort in the composer
The composer shows one control that reads <harness>: <model> <effort>. Open it to see up to three rows:
| Row | What it does |
|---|---|
| Agent | Chooses the harness (OpenCode or Claude Agent). Shown only when you create a session; the harness is fixed afterwards. See Agent harnesses. |
| Model | Lists the enabled models the chosen harness can run, grouped by provider, with each model's display name and one-line description. |
| Effort | Lists Default plus the efforts the model supports (None, Low, Medium, High, XHigh, Max). Hidden for models without reasoning controls. |
Default effort means the model's own default from the table below. Choosing a specific effort applies it to the prompts you send.
You can change the model and effort for each follow-up; the harness cannot change. The list therefore only contains models the session's harness can run. If a message override names a model the harness cannot run, the prompt is rejected with an error such as Model "openai/gpt-5.5" cannot run on the Claude Agent harness. It is never silently replaced with another model.
Model catalog
Reasoning efforts and defaults below come from the shared model catalog. A model with no reasoning efforts has no Effort row in the composer.
Anthropic
Runs on both harnesses. Credential: ANTHROPIC_API_KEY as a global secret, or on Claude Agent sessions a connected Claude account (see Provider accounts).
| Model ID | Display name | Reasoning efforts | Default effort |
|---|---|---|---|
anthropic/claude-haiku-4-5 | Claude Haiku 4.5 | high, max | max |
anthropic/claude-sonnet-4-5 | Claude Sonnet 4.5 | high, max | max |
anthropic/claude-sonnet-4-6 | Claude Sonnet 4.6 | low, medium, high, max | high |
anthropic/claude-sonnet-5 | Claude Sonnet 5 | low, medium, high, xhigh, max | high |
anthropic/claude-opus-4-5 | Claude Opus 4.5 | high, max | max |
anthropic/claude-opus-4-6 | Claude Opus 4.6 | low, medium, high, max | high |
anthropic/claude-opus-4-7 | Claude Opus 4.7 | low, medium, high, xhigh, max | high |
anthropic/claude-opus-4-8 | Claude Opus 4.8 | low, medium, high, xhigh, max | high |
anthropic/claude-opus-5 | Claude Opus 5 | low, medium, high, xhigh, max | high |
anthropic/claude-opus-5-5 | Claude Opus 5.5 | low, medium, high, xhigh, max | high |
anthropic/claude-fable-5 | Claude Fable 5 | low, medium, high, xhigh, max | high |
anthropic/claude-fable-5-1 | Claude Fable 5.1 | low, medium, high, xhigh, max | high |
OpenAI
Runs on OpenCode only. Credential: a connected ChatGPT account, or OPENAI_API_KEY in the session's secret scope when the session uses API-key mode.
| Model ID | Display name | Reasoning efforts | Default effort |
|---|---|---|---|
openai/gpt-5.4 | GPT 5.4 | none, low, medium, high, xhigh | Not set |
openai/gpt-5.5 | GPT 5.5 | none, low, medium, high, xhigh | Not set |
openai/gpt-5.6-sol | GPT 5.6 Sol | none, low, medium, high, xhigh | medium |
openai/gpt-5.6-terra | GPT 5.6 Terra | none, low, medium, high, xhigh | medium |
openai/gpt-5.6-luna | GPT 5.6 Luna | none, low, medium, high, xhigh, max | medium |
openai/gpt-6-astra | GPT-6 Astra | low, medium, high, xhigh, max | medium |
openai/gpt-6-sol | GPT-6 Sol | none, low, medium, high, xhigh, max | medium |
openai/gpt-6-luna | GPT-6 Luna | none, low, medium, high, xhigh, max | medium |
openai/gpt-5.3-codex | GPT 5.3 Codex | low, medium, high, xhigh | high |
openai/gpt-5.3-codex-spark | GPT 5.3 Codex Spark | low, medium, high, xhigh | high |
"Not set" means the composer shows Default and sends no effort unless you choose one.
xAI / SuperGrok
Runs on OpenCode only. Opt-in group. Credential: a connected SuperGrok account, or XAI_API_KEY in API-key mode. Entitlement to a given Grok model is decided by xAI for the connected account.
| Model ID | Display name | Reasoning efforts | Default effort |
|---|---|---|---|
xai/grok-4.5 | Grok 4.5 | low, medium, high | high |
xai/grok-4.6 | Grok 4.6 | low, medium, high, xhigh | high |
xai/grok-4.7 | Grok 4.7 | low, medium, high, xhigh | high |
xai/grok-build-0.1 | Grok Build 0.1 | Not configurable | N/A |
OpenCode Zen
Runs on OpenCode only. Opt-in group. Credential: OPENCODE_API_KEY as a global or repository secret. Zen is billed per token.
| Model ID | Display name | Reasoning efforts | Default effort |
|---|---|---|---|
opencode/kimi-k2.5 | Kimi K2.5 | Not supported | N/A |
opencode/kimi-k2.6 | Kimi K2.6 | Not supported | N/A |
opencode/kimi-k3 | Kimi K3 | Not supported | N/A |
opencode/minimax-m2.5 | MiniMax M2.5 | Not supported | N/A |
opencode/qwen3.7-max | Qwen3.7 Max | Not supported | N/A |
opencode/glm-5 | GLM 5 | Not supported | N/A |
opencode/glm-5.1 | GLM 5.1 | Not supported | N/A |
opencode/glm-5.2 | GLM 5.2 | Not supported | N/A |
OpenCode Go
Runs on OpenCode only. Opt-in group. Credential: the same OPENCODE_API_KEY secret as Zen, but the key must carry an active OpenCode Go subscription. Go bills against the subscription's rolling allowance instead of per token, so a key without a Go subscription fails on opencode-go/* models while opencode/* models keep working.
Rolling allowance and unattended sessions
Go usage is capped on three rolling windows: 20% of the monthly allowance per 5 hours, 50% per week, and 100% per month. Allowances differ per model. An unattended session pinned to a Go model can exhaust its window and fail mid-run, so keep Go models off Slack, GitHub, Linear, and automation launches unless you accept that.
| Model ID | Display name | Reasoning efforts | Default effort |
|---|---|---|---|
opencode-go/grok-4.6 | Grok 4.6 | Not supported | N/A |
opencode-go/gpt-5.6-luna | GPT 5.6 Luna | Not supported | N/A |
opencode-go/glm-5.3-flash | GLM 5.3 Flash | Not supported | N/A |
opencode-go/glm-5.3 | GLM 5.3 | Not supported | N/A |
opencode-go/glm-5.2 | GLM 5.2 | Not supported | N/A |
opencode-go/glm-5.1 | GLM 5.1 | Not supported | N/A |
opencode-go/kimi-k3 | Kimi K3 | Not supported | N/A |
opencode-go/kimi-k2.7-code | Kimi K2.7 Code | Not supported | N/A |
opencode-go/kimi-k2.6 | Kimi K2.6 | Not supported | N/A |
opencode-go/longcat-2.0 | LongCat 2.0 | Not supported | N/A |
opencode-go/deepseek-v4.1-flash | DeepSeek V4.1 Flash | Not supported | N/A |
opencode-go/deepseek-v4-pro | DeepSeek V4 Pro | Not supported | N/A |
opencode-go/deepseek-v4-flash | DeepSeek V4 Flash | Not supported | N/A |
opencode-go/deepseek-v4-flash-vision-exp | DeepSeek V4 Flash Vision Exp | Not supported | N/A |
opencode-go/mimo-v2.5 | MiMo V2.5 | Not supported | N/A |
opencode-go/mimo-v2.5-pro | MiMo V2.5 Pro | Not supported | N/A |
opencode-go/minimax-m3 | MiniMax M3 | Not supported | N/A |
opencode-go/minimax-m2.7 | MiniMax M2.7 | Not supported | N/A |
opencode-go/muse-spark-1.3-contributor | Muse Spark 1.3 Contributor | Not supported | N/A |
opencode-go/muse-spark-1.2-contributor | Muse Spark 1.2 Contributor | Not supported | N/A |
opencode-go/qwen3.8-max | Qwen3.8 Max | Not supported | N/A |
opencode-go/qwen3.8-flash | Qwen3.8 Flash | Not supported | N/A |
opencode-go/qwen3.7-max | Qwen3.7 Max | Not supported | N/A |
opencode-go/qwen3.7-plus | Qwen3.7 Plus | Not supported | N/A |
opencode-go/qwen3.6-plus | Qwen3.6 Plus | Not supported | N/A |
opencode-go/hy4-preview | Hy4 Preview | Not supported | N/A |
opencode-go/hy3 | Hy3 | Not supported | N/A |
Go also publishes minimax-m2.5, but the pinned OpenCode release does not resolve it on the Go gateway, so it is left out of the catalog. Use opencode/minimax-m2.5 on Zen instead.
Z.AI Coding Plan
Runs on OpenCode only. Opt-in group. Credential: ZHIPU_API_KEY as a global or repository secret.
| Model ID | Display name | Reasoning efforts | Default effort |
|---|---|---|---|
zai-coding-plan/glm-5.2 | GLM 5.2 | Not supported | N/A |
zai-coding-plan/glm-5.3 | GLM 5.3 | Not supported | N/A |
DeepSeek
Runs on OpenCode only. Opt-in group. Credential: DEEPSEEK_API_KEY as a global or repository secret.
| Model ID | Display name | Reasoning efforts | Default effort |
|---|---|---|---|
deepseek/deepseek-v4-flash | DeepSeek V4 Flash | Not supported | N/A |
deepseek/deepseek-v4-pro | DeepSeek V4 Pro | Not supported | N/A |
The same model behind several gateways
Some models appear more than once because they are reachable through more than one provider, each billed against a different credential. opencode-go/grok-4.6 and xai/grok-4.6 are the same Grok model behind two gateways. opencode-go/glm-5.2, opencode/glm-5.2, and zai-coding-plan/glm-5.2 are the same GLM model behind three. Pick the entry whose billing you want.
Defaults for integrations and automations
Sessions started by Slack, GitHub, and Linear always run on the OpenCode harness. Automations run on the harness chosen in the automation editor (OpenCode by default). Each surface picks its model from these settings:
| Surface | Where the default lives |
|---|---|
| GitHub | Settings › Integrations › GitHub: Default model and Default reasoning effort, plus per-repository Repository Overrides. Falls back to the deployment default model. |
| Linear | Settings › Integrations › Linear: Default model and effort and Repository Overrides. A model:* issue label wins when Allow model labels is on, then a Linear user preference (when allowed), then the repository override or global default, then the deployment default. |
| Slack | Each Slack user sets Model and Reasoning effort in the app's Home tab. A request can start with !model and !reasoning flags to override them; on a request that starts a session the flags become that session's defaults. |
| Automations | The automation editor picks a model and effort per automation. The list is limited to enabled models the automation's harness can run. |
All of these selectors show only models enabled in Settings › Models. Reasoning values must be ones the selected model supports.
@OpenInspect !model anthropic/claude-sonnet-4-6 !reasoning max investigate the flaky testTroubleshooting
Next steps
Sandbox providers
Compare the five sandbox providers OpenInspect can run sessions on and see which behaviors differ from a user's point of view.
Agent harnesses: OpenCode and Claude Agent
Compare the two agent harnesses, choose one when you create a session, and learn how Claude Agent sessions authenticate and run.