Reduce model loading time #43
Merged
Conversation
This file contains 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
maekawatoshiki
commented
Mar 12, 2023
main.cpp
Outdated
| @@ -496,6 +501,8 @@ bool llama_model_load(const std::string & fname, llama_model & model, gpt_vocab | |||
| fin.close(); | |||
| } | |||
|
|
|||
| free(f_buf); | |||
There was a problem hiding this comment.
f_buf will not be free if this function returns earlier, but I think it does not matter since it's a small amount of memory :)
ggerganov
requested changes
Mar 12, 2023
|
Thank you for your review. Fixed as you mentioned. |
ggerganov
approved these changes
Mar 13, 2023
rooprob
pushed a commit
to rooprob/llama.cpp
that referenced
this pull request
Aug 2, 2023
Speed up rmsnorm by using sqrtf/expf
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Outdated suggestions cannot be applied.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.
Hello!
I noticed that the model loader is not using buffered IO, so I added a piece of code for buffering.
I measured the loading time only for llama 7B on my M1 Pro Macbook, but it reduced the time from 1316ms to 749ms.