anomalyco / anomalyco/opencode

OpenCode Zen muse-spark-1.2-contributor-free suddenly returns HTTP 500 with OpenAI-compatible provider

Open
#44,847 2 comments 0 reactions 1 assignee View on GitHub

@fwang is already working on this.

Since Aug 25, 2026.

Dominant language
TypeScript
Stars
209k
Forks
27.5k
PR merge metrics
PR metrics pending

Description

Description

OpenCode Zen muse-spark-1.2-contributor-free suddenly returns HTTP 500 with OpenAI-compatible provider

Description

I was successfully using muse-spark-1.2-contributor-free with Cline in VS Code through the OpenCode Zen API a few days ago. It suddenly stopped working and now Cline reports:

Internal server error

Other OpenCode Zen models continue to work correctly with the exact same Cline configuration, API key, and Base URL.

The interesting part is that the model itself is healthy and works correctly through OpenCode's Responses API.

Environment

  • Cline: 4.1.15
  • VS Code
  • macOS
  • Provider: OpenAI Compatible
  • Base URL: https://opencode.ai/zen/v1
  • Model: muse-spark-1.2-contributor-free

Current Cline configuration

Provider: OpenAI Compatible
Base URL: https://opencode.ai/zen/v1
Model: muse-spark-1.2-contributor-free

Reproduction

The model is present in the Zen model list:

curl -i https://opencode.ai/zen/v1/models

This returns HTTP 200 and includes:

{
  "id": "muse-spark-1.2-contributor-free",
  "object": "model",
  "owned_by": "opencode"
}
Test 1: Chat Completions API

Using the same API key:

curl -i https://opencode.ai/zen/v1/chat/completions \
  -H "Authorization: Bearer $OPENCODE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "muse-spark-1.2-contributor-free",
    "messages": [
      {
        "role": "user",
        "content": "Say hello"
      }
    ]
  }'

Result:

HTTP/2 500

Response:

{
  "type": "error",
  "error": {
    "type": "error",
    "message": "Internal server error"
  }
}
Test 2: Responses API

Using the exact same API key and model:

curl -i https://opencode.ai/zen/v1/responses \
  -H "Authorization: Bearer $OPENCODE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "muse-spark-1.2-contributor-free",
    "input": "Say hello"
  }'

Result:

HTTP/2 200

The model successfully returns:

{
  "object": "response",
  "status": "completed",
  "model": "muse-spark-1.2-contributor-free",
  "error": null,
  "cost": "0"
}

and generates:

Hello! 👋 How can I help you today?

Expected behavior

Since this model worked with Cline through OpenCode Zen a few days ago, I would expect the Cline OpenAI-compatible configuration to continue working, or for Cline to provide a way to use the OpenAI Responses API when a custom provider/model requires it.

Actual behavior

Cline sends the request through the OpenAI-compatible Chat Completions endpoint, which currently results in:

POST https://opencode.ai/zen/v1/chat/completions
→ HTTP 500

while the same model works through:

POST https://opencode.ai/zen/v1/responses
→ HTTP 200

Why this seems significant

This does not appear to be an API-key, network, or model-availability problem.

The same API key can successfully:

  • retrieve the Zen model list;
  • invoke muse-spark-1.2-contributor-free through /responses;
  • use other Zen models through Cline.

The failure appears specific to the interaction between the Cline OpenAI-compatible provider and the current Zen handling of muse-spark-1.2-contributor-free.

Request

Could Cline maintainers confirm whether the OpenAI-compatible provider is expected to support models that require the OpenAI Responses API?

If possible, it would be useful to have either:

  1. automatic detection/use of /responses for models that require it; or
  2. a provider configuration option allowing the user to select Chat Completions vs Responses.

At minimum, a more descriptive error than Internal server error would make this much easier to diagnose.

Additional context

This worked for me with the same model and Cline configuration a few days ago. The issue appeared recently without me intentionally changing the Base URL, API key, or model ID.

OpenCode itself continues to use the model successfully.

Cline version 4.1.13 and later also include changes/fixes related to custom OpenAI-compatible models, so I am including the exact Cline version (4.1.15) in case this is related to a recent provider/request-handling change.

I can provide additional Cline debug logs or request details if needed.

Plugins

No response

OpenCode version

No response

Steps to reproduce

No response

Screenshot and/or share link

No response

Operating System

No response

Terminal

No response

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.