[Feature]: Contribute T5 model to vLLM

[shivance]

🚀 The feature, motivation and pitch

Hi vLLM team,
As you already know, T5 is a perfect model and as mentioned in #7366, the project intends to add support for T5. I want to help the cause and contribute T5 to the project.

As this is my first contribution to vLLM project I am quite pumped up as well as eager to learn from you all.

Alternatives

No response

Additional context

No response

Before submitting a new issue...

• Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

[robertgshaw2-redhat]

@shivance thanks for your interest!

See: #7366 for a detailed discussion of enc-dec support. We would definitely welcome help on T5. For T5, the key is that we need an attention implementation with the custom bias

• [RFC]: Encoder/decoder models & feature compatibility #7366 (comment)

[github-actions]

This issue has been automatically marked as stale because it has not had any activity within 90 days. It will be automatically closed if no further activity occurs within 30 days. Leave a comment if you feel this issue should remain open. Thank you!

[github-actions]

This issue has been automatically closed due to inactivity. Please feel free to reopen if you feel it is still relevant. Thank you!