Check this out. Couldnt try it[Bug]: GENERATION_FAILED (fp: 78910818)

[luis782006]

Description

I doesnt work. Not even with your example.
Error invoking remote method 'codesign:v1:generate': CodesignError: 404 model 'models/gemini-2.5-pro' not found.

Model was connected. At least I was told about that when I try connection.
Looks great app.

Steps to reproduce

Expected behavior

To start to do someting

Actual behavior

Error invoking remote method 'codesign:v1:generate': CodesignError: 404 model 'models/gemini-2.5-pro' not found

open-codesign version

0.1.4

Platform

Windows

OS version

Windows 11 (10.0.26200)

Model provider (if relevant)

None

Error code

GENERATION_FAILED

Diagnostics bundle

Bundle saved locally at ~\Downloads\open-codesign-diagnostics-2026-05-02T01-54-12.zip. Attach it to this issue after submitting.

Logs

[2026-05-01 22:53:53.710] [info]   (main:ipc) [generate] step=send_request {
  runId: 'monoubko-mmzcsz',
  generationId: 'monoubko-mmzcsz',
  provider: 'ollama',
  modelId: 'models/gemini-2.5-pro'
}
[2026-05-01 22:53:53.711] [info]   (main:ipc) agent.turn_start { runId: 'monoubko-mmzcsz', generationId: 'monoubko-mmzcsz' }
[2026-05-01 22:53:53.712] [info]   (main:ipc) [context-prune] step=caps {
  runId: 'monoubko-mmzcsz',
  generationId: 'monoubko-mmzcsz',
  messages: 3,
  before: 685,
  after: 685,
  textLimit: 8192,
  toolInputLimit: 24576,
  toolResultLimit: 8192,
  window: 3
}
[2026-05-01 22:53:53.725] [info]   (main:ipc) agent.turn_end {
  runId: 'monoubko-mmzcsz',
  generationId: 'monoubko-mmzcsz',
  deltas: 0,
  tools: 0
}
[2026-05-01 22:53:53.726] [info]   (main:ipc) agent.end { runId: 'monoubko-mmzcsz', generationId: 'monoubko-mmzcsz' }
[2026-05-01 22:53:53.727] [error]  (main:ipc) [generate] step=send_request.fail {
  runId: 'monoubko-mmzcsz',
  generationId: 'monoubko-mmzcsz',
  provider: 'ollama',
  modelId: 'models/gemini-2.5-pro',
  ms: 16,
  stopReason: 'error'
}
[2026-05-01 22:53:53.728] [error]  (main:ipc) generate.fail {
  runId: 'monoubko-mmzcsz',
  generationId: 'monoubko-mmzcsz',
  ms: 23,
  provider: 'ollama',
  modelId: 'models/gemini-2.5-pro',
  baseUrl: '[url omitted]',
  status: 404,
  message: "404 model 'models/gemini-2.5-pro' not found",
  code: 'PROVIDER_ERROR'
}
[2026-05-01 22:53:53.736] [error]  (renderer:console) [store] applyGenerateError {"generationId":"monoubko-mmzcsz","designId":"7e5e3c5a-4e8d-49d5-9d51-6917d426592e","message":"Error invoking remote method 'codesign:v1:generate': CodesignError: 404 model 'models/gemini-2.5-pro' not found"} {}
[2026-05-01 22:53:53.738] [info]   (chat-messages-ipc) chat.append {
  designId: '7e5e3c5a-4e8d-49d5-9d51-6917d426592e',
  seq: 5,
  kind: 'error'
}

[github-actions]

Thank you for reporting this issue! The error 404 model 'models/gemini-2.5-pro' not found with provider ollama indicates a model naming mismatch.

Likely cause: Ollama model IDs should not include the models/ prefix. That prefix is used by Google Gemini's API, but Ollama expects the exact model name as shown by ollama list (e.g., gemma2:2b, llama3.2, or gemini2.5-pro if you've pulled a model with that name).

Check:

• Open a terminal and run ollama list to see the exact names of your installed models.
• In Open CoDesign's settings, set the model ID to the name exactly as listed (e.g., gemini2.5-pro if that's the tag you pulled).

Relevant code: The app passes the modelId directly to the Ollama API (packages/providers/src/ollama.ts). The models/ prefix is not stripped for Ollama, so it sends the full string causing a 404.

Need more info:

• Could you run ollama list and share which model names appear?
• Did you manually type models/gemini-2.5-pro or was it suggested by the UI?

Thanks again for trying Open CoDesign! Let me know if the fix works.

---

Open-CoDesign Bot

[Sun-sunshine06]

Closing as fixed by #315 (main commit 768dc8b). Errors involving a models/... model id on 400/404/422 are now classified as model-id-shape, including no-body gateway responses, and the UI can offer the one-click action to remove the unsupported models/ prefix.

The upcoming v0.2 packaged release will include this fix. If the same model still fails after applying the suggested model-id fix, please open a fresh report with the new diagnostics bundle.