OpenCode Zen and Go Catalogs via OpenClaw
Learn how to use OpenCode's Zen and Go hosted catalogs through OpenClaw, including setup, model defaults, and verification. Essential for developers integrating OpenCode models.
Read this when
- You want OpenCode-hosted model access
- You want to pick between the Zen and Go catalogs
OpenCode provides two hosted catalogs through OpenClaw:
| Catalog | Prefix | Runtime provider |
|---|---|---|
| Zen | opencode/... | opencode |
| Go | opencode-go/... | opencode-go |
The two catalogs rely on the identical OpenCode API key setup (OPENCODE_API_KEY,
alias OPENCODE_ZEN_API_KEY). A separate paid subscription remains necessary for Go; holding a Zen key alone does not automatically unlock Go. OpenClaw maintains distinct runtime provider ids so that upstream per-model routing continues to function properly.
Getting started
Zen catalog
Best for: the curated OpenCode multi-model proxy (Claude, GPT, Gemini, GLM, DeepSeek, Kimi, MiniMax, Qwen).
Run onboarding
openclaw onboard --auth-choice opencode-zen
The key can also be supplied directly:
openclaw onboard --opencode-zen-api-key "$OPENCODE_API_KEY"
Set a Zen model as the default
openclaw config set agents.defaults.model.primary "opencode/gpt-5.6-sol"
Verify models are available
openclaw models list --provider opencode
Go catalog
Best for: the separately subscribed Go lineup across DeepSeek, GLM, GPT, Grok, Hy3, Kimi, MiMo, MiniMax, and Qwen.
Use the bundled Go catalog
For this release, OpenClaw ships with OpenCode Go already included, so no extra plugin installation or Gateway restart is needed.
Run onboarding
openclaw onboard --auth-choice opencode-go
Alternatively, pass the key straight through:
openclaw onboard --opencode-go-api-key "$OPENCODE_API_KEY"
Set a Go model as the default
openclaw config set agents.defaults.model.primary "opencode-go/kimi-k3"
Verify models are available
openclaw models list --provider opencode-go
Config example
{
env: { vars: { OPENCODE_API_KEY: "sk-..." } },
agents: { defaults: { model: { primary: "opencode/gpt-5.6-sol" } } },
}
Provider catalogs
Zen
| Property | Value |
|---|---|
| Runtime provider | opencode |
| Example models | opencode/gpt-5.6-sol, opencode/kimi-k3, opencode/deepseek-v4-flash |
Execute openclaw models list --provider opencode to see the currently active list.
Model availability and promotional routes may shift independently of OpenClaw.
Live discovery merges the models accessible to your OpenCode account with
authoritative model metadata sourced from https://models.opencode.ai/api.json.
OpenClaw retrieves and caches that catalog only when OpenCode Zen or Go is
enabled or explicitly chosen with OpenCode credentials; startup and
unrelated providers never trigger the download. New upstream models appear
without an OpenClaw update as long as their metadata specifies a supported transport
on the trusted OpenCode endpoint. A response scoped to a key can omit models
unavailable to that workspace. Metadata and lifecycle status refresh in tandem;
deprecated models are left out of active discovery and its offline fallback.
Deprecated explicit refs stay resolvable for existing configurations but are
no longer presented as current recommendations.
Go
| Property | Value |
|---|---|
| Runtime provider | opencode-go |
| Example models | opencode-go/kimi-k3, opencode-go/deepseek-v4-flash, opencode-go/qwen3.8-max |
Refer to OpenCode Go for discovery, routing, and access requirements. The model-list endpoint for Go advertises its general lineup; a model being listed does not guarantee your account can execute it.
Advanced configuration
API key aliases
As an alias for OPENCODE_API_KEY, OPENCODE_ZEN_API_KEY is also recognized.
Shared credentials
During setup, a single OpenCode key entry can store credentials for both runtime providers. That action does not establish a Go subscription or confer Go entitlement; subscribe to Go in the OpenCode console before using it.
Getting an API key
Set up an OpenCode account and obtain an API key at opencode.ai/auth. Billing and catalog availability are handled from the OpenCode dashboard.
Gemini replay behavior
OpenCode refs backed by Gemini remain on the proxy-Gemini path, so OpenClaw applies Gemini thought-signature sanitation there without enabling native Gemini replay validation or bootstrap rewrites.
Non-Gemini replay behavior
Non-Gemini OpenCode refs retain the minimal OpenAI-compatible replay policy.
Pricing and privacy
Billing, retention, and training policies differ by model. Review the current OpenCode Zen pricing and policy before picking a route. Free models may represent temporary feedback programs.
Related
-
OpenCode Go, Discovery and access requirements for the Go catalog.
-
Model selection, Choosing providers, model refs, and failover behavior.
-
Configuration reference, Full config reference for agents, models, and providers.