【Inference Optimize】DeepSeek-v3 model inference performance optimization #3455
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thanks for your contribution! |
yuanlehome
reviewed
Aug 18, 2025
Comment on lines
+157
to
+171
| if self.flash_attn_func is None: | ||
| prop = paddle.device.cuda.get_device_properties() | ||
| cc = prop.major * 10 + prop.minor | ||
| is_current_sm_supported = cc >= 90 | ||
| is_paddle_supported = any(num >= 90 for num in paddle.version.cuda_archs()) | ||
| if is_current_sm_supported and is_paddle_supported: | ||
| self.flash_attn_func = flash_attention_v3_varlen | ||
| print("The current platform supports Flash Attention V3.") | ||
| self.flash_attn_kwargs = {"softmax_scale": self.attn_softmax_scale} | ||
| else: | ||
| self.flash_attn_func = flash_attn_unpadded | ||
| self.flash_attn_kwargs = {"scale": self.attn_softmax_scale, "training": False} | ||
| print( | ||
| "The current platform does not support Flash Attention V3, so Flash Attention V2 will be used instead." | ||
| ) |
There was a problem hiding this comment.
可以把这一块代码从各个派生类中挪到基类里去吗？这样只需要写一次
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
flash_attention_v3_varlen端到端性能提升10%左右。