Model not deploying previously it was deployed #2114
Description
Describe the bug
Hi as I tried to push the custom finetune orpheus tts model on basten using mig h100 it worked, I did the inference it was working pretty well but once I deactivated and made active again it got stuck in
Build was successful. Deploy task is starting.
Configuring Resources to match user provided values
Requesting 7625 millicpus
Requesting 58000 MiB of memory
Requesting 1 h100-40gb GPU
Creating the Baseten Inference Service.
it same for the new workspace I did for the model. it was working fine previously.
To Reproduce
I used this command to push which was working truss push Orpheus-3b-main-upload --publish
Expected behavior
above command gave this output
Model is marked as force-legacy-api-non-openai-compatible. This model will not be compatible with OpenAI clients directly.
This is the deprecated legacy behavior, please update the tag to openai-compatible.
TRT-LLM model is using the deprecated model_metadata['tags']=force-legacy-api-non-openai-compatible tag. While this is
currently working, please consider migrating to openai-compatible tag and request schema.
Model is marked as force-legacy-api-non-openai-compatible. This model will not be compatible with OpenAI clients directly.
This is the deprecated legacy behavior, please update the tag to openai-compatible.
Deploying truss using H100MIG instance type.
✨ Model Orpheus-3b shona redefined accent version 1 was successfully pushed ✨
🪵 View logs for your deployment at https://app.baseten.co/models/4q9x/logs/32pz0
(eleven) shekharmeena@Shekhars-MacBook-Pro data collections elevenlabs %
Screenshots/Logs
If applicable, add screenshots or logs to help explain your problem.
Desktop (please complete the following information):
- OS: [e.g. macOS]
- Environment versions [e.g. python versions, framework/package versions]
- Truss version [e.g. 22]
Additional context
Add any other context about the problem here.
