Skip to content

Cerebras provider zai-glm-4.7 has 400 error after first message #25

@ltanedo

Description

@ltanedo

Pre-submit Checks

Describe the bug

When using Cerebras Provider with zai-glm-4.7, all messages after first message have a 400 bad request error.

Request failed with error: Other(BYOP stream error: 网络/流式请求失败: Web stream error for model 'zai-glm-4.7 (adapter: OpenAI)'.
Cause: HTTP error.
Status: 400 Bad Request Bad Request
Body: {"message":"messages.4.assistant.reasoning_content: property 'messages.4.assistant.reasoning_content' is unsupported","type":"invalid_request_error","param":"validation_error","code":"wrong_api_format"})

Attempting to resume conversation...

Possibly, the provider code is including "reasonning_content" but Cerebras' aoi doesn't support it

To reproduce

  1. Setup Cerebras Provider with proper api key
  2. Send first message using zai-glm-4.7 model
  3. Send second message, will result in 400 error code on api

Expected behavior

No error when chatting

Screenshots, videos, and logs

Image

Operating system (OS)

Windows

Operating system and version

Windows 11

Shell Version

Powershell

Current Warp version

v2026.05.03.preview.4

Regression

No, this bug or issue has existed throughout my experience using Warp

Recent working Warp date

No response

Additional context

No response

Does this block you from using Warp daily?

No

Is this an issue only in Warp?

Yes, I confirmed that this only happens in Warp, not other terminals.

Warp Internal (ignore): linear-label:b9d78064-c89e-4973-b153-5178a31ee54e

None

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't working

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions