Skip to content

Support deepseek v3 #9835

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
wants to merge 67 commits into
base: develop
Choose a base branch
from
Open

Support deepseek v3 #9835

wants to merge 67 commits into from

Conversation

zhoutianzi666
Copy link
Contributor

Before submitting

  • Lint code. If there are lint issues, please format the code first.
# Install and register `pre-commit` in the project folder
pip install pre-commit && pre-commit install

# Process previous code files separately
pre-commit run --file XXXX.py
  • Add test cases into tests folder. If there are codecov issues, please add tests cases first.

PR types

PR changes

Description

yuanlehome and others added 30 commits January 13, 2025 17:11
@yuanlehome
support deepseek-v3
f3c1336
@lizhenyun01
support head_dim=192,256 for append_attn c16
09f94b9
@yuanlehome
update 0113
dbfd416
@yuanlehome
Merge pull request PaddlePaddle#2 from lizhenyun01/append_attn_headdim
f1894f2 
support  append_attn c16 for deep-seek-v3
@yuanlehome
attention run
e475e3b
@yuanlehome
refine code
c9b33cd
@yuanlehome
add softmax_scale
dc92a3b
@yuanlehome
support weight_only_int8
6d59b51
@yuanlehome
refine code
07c6bb2
@yuanlehome
support tp
73ad324
@yuanlehome
delete test_append_attn
cbe3623
@yuanlehome
add splited fused_moe from ziyuan
f5d5d24
@lizhenyun01
fix repe for deepseek-v3
9657580
@yuanlehome
add deepseek-v3 class
feabdb8
@yuanlehome
Merge pull request PaddlePaddle#4 from lizhenyun01/fix_rope
acc025f 
Fix rope&fix precision
@yuanlehome
fix wint8 precision and refine code
583e17f
@yuanlehome
fix wint4, big diff
69823b6
@yuanlehome
add e_score_correction_bias
2877f2c
@yuanlehome
fix head_dim
9fd7a1d
@yuanlehome
fix v3 verify
abb63f0
@yuanlehome
Merge branch 'develop' of https://github.com/PaddlePaddle/PaddleNLP i…
ac0b93e 
…nto support-deepseek-v3
@yuanlehome
fix d2s
47fd938
@yuanlehome
fix v3 verify
57f6dce
@yuanlehome
support qk_head_dim != v_head_dim
f1cac13
@yuanlehome
fix wint8 v_head_dim
271be86
@lizhenyun01
fix rope
d36f5bb
@yuanlehome
Merge pull request PaddlePaddle#7 from lizhenyun01/fix_rope
d6e6dff 
fix rope
@yuanlehome
fix qwen2
d9283ca
@yuanlehome
mla use position_ids only
239564b
@yuanlehome
remove control flow
f86e4f7
@paddle-bot Paddle-bot
Copy link

paddle-bot bot commented Feb 10, 2025

Thanks for your contribution!

@codecov Codecov
Copy link

codecov bot commented Feb 13, 2025 •
edited
Loading

Codecov Report

Attention: Patch coverage is 0.64103% with 155 lines in your changes missing coverage. Please review.

Project coverage is 49.89%. Comparing base (ee3de26) to head (81bd9d6).
Report is 224 commits behind head on develop.

Files with missing lines Patch % Lines
...erimental/transformers/fused_transformer_layers.py 0.00% 122 Missing ⚠️
...lp/experimental/transformers/qwen2_moe/modeling.py 0.00% 19 Missing ⚠️
paddlenlp/transformers/conversion_utils.py 11.11% 8 Missing ⚠️
...enlp/experimental/transformers/generation_utils.py 0.00% 3 Missing ⚠️
paddlenlp/trl/llm_utils.py 0.00% 3 Missing ⚠️

❌ Your patch check has failed because the patch coverage (0.64%) is below the target coverage (80.00%). You can increase the patch coverage or adjust the target coverage.
❌ Your project check has failed because the head coverage (49.89%) is below the target coverage (58.00%). You can increase the head coverage or adjust the target coverage.

Additional details and impacted files 
@@             Coverage Diff             @@
##           develop    #9835      +/-   ##
===========================================
- Coverage    49.95%   49.89%   -0.06%     
===========================================
  Files          757      757              
  Lines       122540   122691     +151     
===========================================
+ Hits         61215    61219       +4     
- Misses       61325    61472     +147

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

@CLAassistant
Copy link

CLAassistant commented Mar 13, 2025 •
edited
Loading

CLA assistant check
Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution.
4 out of 5 committers have signed the CLA.

✅ yuanlehome
✅ lizhenyun01
✅ ckl117
✅ zhoutianzi666
❌ root

root seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
You have signed the CLA already but the status is still pending? Let us recheck it.

@github-actions GitHub Actions
Copy link

This Pull Request is stale because it has been open for 60 days with no activity. 当前Pull Request 60天内无活动，被标记为stale。

@github-actions github-actions bot added the stale label May 24, 2025
@github-actions github-actions bot removed the stale label Aug 3, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

5 participants

@zhoutianzi666 @CLAassistant @yuanlehome @lizhenyun01 @ckl117