the model loading requires that a text field for a model name is filled, despite there being dropdowns lower down in the settings. that UI makes no sense, the model list loading should work without a model and model selection fields should be, well, selections.
Use my localhost endpoint for model endpoint: http://127.0.0.1:9000/v1 with api key "dummy" and when selecting any model, use the 9B Qwen model.
The local AI endpoint has the same problem, but it doesn't even have an API Key field that could be filled, so fails if the local AI provider (like oMLX) requires an API key.
the model loading requires that a text field for a model name is filled, despite there being dropdowns lower down in the settings. that UI makes no sense, the model list loading should work without a model and model selection fields should be, well, selections.
Use my localhost endpoint for model endpoint: http://127.0.0.1:9000/v1 with api key "dummy" and when selecting any model, use the 9B Qwen model.
The local AI endpoint has the same problem, but it doesn't even have an API Key field that could be filled, so fails if the local AI provider (like oMLX) requires an API key.
hugo
added the bug label 2026-07-19 15:05:19 +00:00
Fixed in commit 2d6fcaf. Model discovery now requires only the endpoint URL and applicable credentials, not a preselected model. Model fields are selections populated by discovery, and local endpoints now support an optional API key.
Fixed in commit 2d6fcaf. Model discovery now requires only the endpoint URL and applicable credentials, not a preselected model. Model fields are selections populated by discovery, and local endpoints now support an optional API key.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
the model loading requires that a text field for a model name is filled, despite there being dropdowns lower down in the settings. that UI makes no sense, the model list loading should work without a model and model selection fields should be, well, selections.
Use my localhost endpoint for model endpoint: http://127.0.0.1:9000/v1 with api key "dummy" and when selecting any model, use the 9B Qwen model.
The local AI endpoint has the same problem, but it doesn't even have an API Key field that could be filled, so fails if the local AI provider (like oMLX) requires an API key.
Fixed in commit
2d6fcaf. Model discovery now requires only the endpoint URL and applicable credentials, not a preselected model. Model fields are selections populated by discovery, and local endpoints now support an optional API key.