[Feature]: specify output to use for custom models
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 3.2k
- Forks
- 249
- Avg merge
- 4d 8h
- Merged PRs (30d)
- 4
Description
What feature would you like to request?
it would be nice to have an optional parameter for custom models to specify which model output to use.
right now i'm using fastembed for testing qwen3, and to make the model fastembed compatible i have to remove the first model output from the graph since i can't specify which model output to use.
this could probably result in small performance gains too, since i assume doing pooling and normalization in optimized onnx (as is sometimes the case with the secondary output) should be slightly faster
Is there any additional information you would like to provide?
No response
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No file or test is named. Start by tracing how fastembed loads custom models and selects model outputs; determine where an optional output selection can fit without changing existing behavior. Done means custom models can choose the intended output and existing models continue to work, with coverage for the new option.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Feature
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100