Transformers version 4.55 upgrade, Update PyTorch to 2.7.0+cpu, Torchvision to 0.22.0+cpu, and Python Requirement to >=3.9 #542
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
quic-mamta
requested review from
ochougul,
quic-amitraj,
quic-hemagnih and
quic-rishinr
as code owners
asmigosw
force-pushed
the
TF_version_4.55_upgrade
branch
from
d36c124 to
a514d36
Compare
quic-mamta
commented
Sep 9, 2025
quic-mamta
force-pushed
the
TF_version_4.55_upgrade
branch
2 times, most recently
from
e15d548 to
3643fee
Compare
quic-mamta
force-pushed
the
TF_version_4.55_upgrade
branch
from
69ec2a4 to
6ad267b
Compare
2 tasks
quic-mamta
changed the title
quic-mamta
force-pushed
the
TF_version_4.55_upgrade
branch
3 times, most recently
from
dd8b38e to
940dfcf
Compare
quic-mamta
force-pushed
the
TF_version_4.55_upgrade
branch
2 times, most recently
from
4f44dd4 to
d8cf0a1
Compare
| # Apply the attention mask | ||
| attn_weights = torch.where(attention_mask, mask_value, attn_weights) | ||
|
|
||
| attn_weights = attn_weights / self.scale_attn |
There was a problem hiding this comment.
Why it has been moved from line 51?
There was a problem hiding this comment.
It was made equivalent to new TF code; they have moved it down since its placement whether at line 50 or 58, won't affect the performance. Should I move it back to 50?
vbaddi
reviewed
Oct 6, 2025
|
|
||
| EXTERNAL_MODELS = { | ||
| "hpcai-tech/grok-1", | ||
| "hpcai-tech/grok-1": { |
quic-mamta
force-pushed
the
TF_version_4.55_upgrade
branch
3 times, most recently
from
8217cb5 to
7bf2298
Compare
Signed-off-by: Asmita Goswami <[email protected]>
Signed-off-by: Asmita Goswami <[email protected]>
Signed-off-by: Asmita Goswami <[email protected]>
Signed-off-by: Asmita Goswami <[email protected]>
Signed-off-by: Asmita Goswami <[email protected]>
Signed-off-by: Mamta Singh <[email protected]>
asmigosw
force-pushed
the
TF_version_4.55_upgrade
branch
from
c9fd8ea to
178650f
Compare
Signed-off-by: Asmita Goswami <[email protected]>
Signed-off-by: Asmita Goswami <[email protected]>
Signed-off-by: Asmita Goswami <[email protected]>
Signed-off-by: Asmita Goswami <[email protected]>
Signed-off-by: Asmita Goswami <[email protected]>
Signed-off-by: Asmita Goswami <[email protected]>
Signed-off-by: Asmita Goswami <[email protected]>
ochougul
reviewed
Oct 13, 2025
Comment on lines
-555
to
-558
| # use local attention mask for ROPE layers | ||
| if self.use_chunked_attention: | ||
| attention_mask = chunk_causal_mask | ||
|
|
ochougul
reviewed
Oct 13, 2025
Comment on lines
722
to
724
| logger.warning( | ||
| "Current version output doesn't match with HF output due to a bug in TF v_4.55. Switch to branch release/v_1.20 for TF match." | ||
| ) |
There was a problem hiding this comment.
put link to this PR in the warning: https://github.com/huggingface/transformers/pull/39501/files#diff-e668ec07f78afdb2cb805d939e47453757f0b9437436cb860fcb7cb2431c9cf5R140
Signed-off-by: Asmita Goswami <[email protected]>
Signed-off-by: Asmita Goswami <[email protected]>
quic-rishinr
approved these changes
Oct 14, 2025
quic-vargupt
pushed a commit
to quic-vargupt/efficient-transformers
that referenced
this pull request
Oct 17, 2025
…vision to 0.22.0+cpu, and Python Requirement to >=3.9 (quic#542) Update Transformers to 4.55.0 Update PyTorch to 2.7.0+cpu Torchvision to 0.22.0+cpu and Python Requirement to >=3.9 Updated modeling files and Cache Utils for transformers 4.55.0 Updated models : 1. codegen 2. falcon 3. gemma 4. gemma2 5. gptj 6. gpt2 7. granite 8. granite_moe 9. grok1 10. llama 11. llama_swiftkv 12. mistral 13. mixtral_moe 14. mpt 15. phi 16. phi3 17. qwen2 18. starcoder2 19. gpt_bigcode 20. internvl 21. llava 22. llava_next 23. whisper 24. gemma3 25. llama4 26. mllama --------- Update Qeff Documentation to indicate vLLM Support in Validated Models Page Signed-off-by: Asmita Goswami <[email protected]> Signed-off-by: Mamta Singh <[email protected]> Co-authored-by: Mamta Singh <[email protected]> Co-authored-by: Asmita Goswami <[email protected]> Signed-off-by: Varun Gupta <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
7 participants
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.
Update Transformers to 4.55.0
Update PyTorch to 2.7.0+cpu
Torchvision to 0.22.0+cpu
and Python Requirement to >=3.9
Updated modeling files and Cache Utils for transformers 4.55.0
Updated models :