NVIDIA / NVIDIA/TensorRT

how to support FlashAttention2?

Open
#4,082 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
C++
Stars
13.4k
Forks
2.4k
Avg merge
5d 3h
Merged PRs (30d)
2

Description

Hi, does the currently transferred trt engine support flash focus2 by default? If it is not supported by default, how should I use fa2 in the output engine?
Thanks,

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by checking the current TensorRT engine support for FlashAttention2 and any existing integration or usage guidance. Determine whether FlashAttention2 is supported by default and, if not, identify the documented steps required to use it in the output engine; the issue is complete when this behavior and procedure are clearly explained.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.