Skip to content

Cerebras Qwen3-Coder Is Deprecated: OpenCode Migration

Cerebras deprecated its public Qwen3-Coder-480B model. Replace the broken OpenCode setup with a current public model or a Qwen provider.

MGMCSA Guru Team August 22, 2026 2 min read
OpenCode migration from the deprecated Cerebras Qwen3-Coder model

The setup originally published at this URL no longer works on Cerebras’s shared public API. Cerebras deprecated qwen-3-coder-480b on November 5, 2025. Keeping the old model in OpenCode can produce a model-not-found response.

Pick the migration path

If you want to stay on the public Cerebras service, list the models available to your account and choose a current production model. Cerebras’s public catalog changes, so copy the ID returned by the API rather than reusing an old article.

If you specifically want Qwen, use Alibaba Model Studio or Requesty. Requesty’s current managed ID is qwen3.8-max, and its live catalog can route that model across several providers.

Update OpenCode for Requesty

Store a Requesty key in REQUESTY_API_KEY, then add an OpenAI-compatible provider:

{
  "$schema": "https://opencode.ai/config.json",
  "provider": {
    "requesty-extra": {
      "npm": "@ai-sdk/openai-compatible",
      "name": "Requesty",
      "options": {
        "baseURL": "https://router.requesty.ai/v1",
        "apiKey": "{env:REQUESTY_API_KEY}"
      },
      "models": {
        "qwen3.8-max": { "name": "Qwen3.8-Max" }
      }
    }
  },
  "model": "requesty-extra/qwen3.8-max"
}

Restart OpenCode, open the model picker, and confirm that Qwen3.8-Max appears. Test a read-only request before approving edits.

Remove the retired ID

Search user and project OpenCode configuration for:

qwen-3-coder-480b

Remove every active reference, including fallback lists. A hidden fallback can keep generating errors after the primary model has been changed.

Cerebras Qwen migration checklist

  • Deprecated qwen-3-coder-480b references removed
  • Current provider and model selected
  • API key stored outside the repository
  • OpenCode restarted
  • Read-only tool request tested

For direct Alibaba use, see Qwen3.8-Max with Claude Code. For current prices, see Qwen3.8-Max pricing.

Frequently asked questions

Is qwen-3-coder-480b still available on the Cerebras public API?

No. Cerebras lists it as deprecated from November 5, 2025. Old OpenCode configurations can now return a model-not-found error.

What should replace it?

Use a current Cerebras public model if speed is the priority, or move the Qwen workflow to Alibaba Model Studio or Requesty with qwen3.8-max.

Can a dedicated Cerebras endpoint still host Qwen?

Cerebras lists Qwen families among its dedicated-endpoint options. That is a separate enterprise deployment and does not make the retired public model ID valid.

Sources & further reading

Official vendor documentation referenced while writing this guide.

MG

MCSA Guru Team

IT & Systems Administration

We are working IT pros and system administrators who spend our days in Windows Server, Microsoft 365, and the wider Microsoft stack. MCSA Guru is where we write down the fixes and walkthroughs we wish we had found the first time.

MCSA Guru provides independent, educational IT guidance. Microsoft, Windows, Windows Server, Microsoft 365, Exchange, and Microsoft Teams are trademarks of Microsoft Corporation; Docker is a trademark of Docker, Inc. MCSA Guru is not affiliated with or endorsed by Microsoft or Docker. Always test changes in a safe environment before applying them in production.

Related guides

Fixing something right now?

Jump straight into the guide library or search for the exact error or task you are dealing with.