Skip to content
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

Sgl project main #10

Merged
merged 99 commits into from
Oct 23, 2024
Merged

Sgl project main #10

merged 99 commits into from
Oct 23, 2024

Conversation

81549361
Copy link
Owner

Motivation

Modifications

Checklist

  • Format your code according to the Contributor Guide.
  • Add unit tests as outlined in the Contributor Guide.
  • Update documentation as needed, including docstrings or example tutorials.

ByronHsu and others added 30 commits October 10, 2024 16:34
@ByronHsu
Fix test_vision_openai_server on CI (sgl-project#1620)
01fdb2f
@HaiShaw
[Performance, hardware] MoE tuning update to AMD MI300x GPUs (sgl-pro…
e11ab79 
…ject#1619)
@kushal34712
Update README.md (sgl-project#1625)
c9e6658
@merrymercy
Update README.md (sgl-project#1629)
b040ed7
@merrymercy
Update README.md
5476cca
@liangan1
Add device support (sgl-project#1607)
8275049
@glen-amd
Nit about the decorator of PortArgs.init_new (sgl-project#1611)
58093b8
@OBJECT907
[Bug] Fix the Image Input of Batch Generation (sgl-project#1579)
b503881
@Abatom
Add the ability to enable and disable the Profiler via HTTP API. (sgl…
bbd72bf 
…-project#1626)
@merrymercy
Fix the correctness test in bench_latency.py when tp > 1 and test_gen…
aba9eae 
…eration_models.py (sgl-project#1631)
@merrymercy @yizhang2077
Add image_token in conversation.py (sgl-project#1632)
f13d86f 
Co-authored-by: yizhang2077 <[email protected]>
@JanumalaAkhilendra
Added a "Back To Top" Button (sgl-project#1633)
81c3327
@merrymercy
Fix constrained decoding (sgl-project#1634)
5d09ca5
@merrymercy
Add back data parallelism (sgl-project#1635)
23cc66f
@merrymercy
Release v0.3.3.post1 (sgl-project#1636)
00c7e63
@ByronHsu
[engine] support async and streaming (sgl-project#1614)
862cd26
@merrymercy
[Fix] Fix the style of test_large_max_new_tokens.py (sgl-project#1638)
dafb6a5
@learninmou
fix missing ignore_eos in v1/chat/completions (sgl-project#1642)
1d9deea
@merrymercy
Fix ignore_eos (sgl-project#1645)
e37cdab
@liangan1
[Feature, Hardware] Enable SGLang on XPU GPUs via PyTorch (sgl-projec…
5d638c9 
…t#1480)
@merrymercy
Fix unit tests and type annotations (sgl-project#1648)
69aa937
@merrymercy
Add an option to disable penalizer (sgl-project#1651)
9da5a60
@pjyi2147
Add get_tokenizer function for Engine class (sgl-project#1653)
31fad29
@merrymercy
Fix the batch_is_full check for jump-forward decoding (sgl-project#1654)
9610fcd
@merrymercy
Simplify the event loop and expose --num-continuous-decode-steps as…
7ee6c25 
… an argument (sgl-project#1652)
@ByronHsu
[doc] Add engine section in backend.md (sgl-project#1656)
c3f2fc5
@Ying1123
[Fix] fix eos trim inconsistency (sgl-project#1650)
4876117
@merrymercy
Add output_ids into ScheduleBatch (sgl-project#1659)
da1ffed
@Ying1123
[Minor] Rename no_eos_trim to no_stop_trim (sgl-project#1661)
2725f8d
@merrymercy
Add a test case to test retract (sgl-project#1662)
869f1c0
zhyncs and others added 28 commits October 19, 2024 20:45
@zhyncs @ispobock
Update vllm to 0.6.3 (sgl-project#1711) (sgl-project#1720)
8bee20f 
Co-authored-by: Ke Bao <[email protected]>
@zhyncs @yizhang2077 @ispobock
Support qwen2 vl model (sgl-project#1721)
cbbc82b 
Co-authored-by: yizhang2077 <[email protected]>
Co-authored-by: ispobock <[email protected]>
@Ying1123
Update README.md (sgl-project#1722)
5c4ce65
@Ying1123
Update README.md
9594627
@merrymercy
Unify the memory pool api and tp worker API (sgl-project#1724)
59cbf47
@merrymercy
Temporarily skip this test_mixed_batch for QWen2VL (sgl-project#1725)
593b19f
@merrymercy
Split the overlapped version of TpModelWorkerClient into a separate f…
b48edff 
…ile (sgl-project#1726)
@yizhang2077
[Bugfix] qwen2vl forward_extend (sgl-project#1727)
554fbf9
@merrymercy
Simplify the usage of device (sgl-project#1734)
e12358d
@merrymercy
Simplify batch result resolution (sgl-project#1735)
b121bc0
@sixsixcoder
Add GLM-4 TextGeneration Model support for SGLang (sgl-project#1736)
45d5af2
@merrymercy
Make token mapping non-blocking in the overlapped mode (sgl-project#1740
cf470fe 
)
@merrymercy
Maintain seq_lens_sum to make more FlashInfer operations non-blocking (…
09603c6 
…sgl-project#1741)
@hnyls2002
Fix prefill oom (sgl-project#1743)
efb099c
@merrymercy
Faster overlap mode scheduler (sgl-project#1738)
7ce3606
@zhyncs
misc: add CODEOWNERS (sgl-project#1737)
e68b9e7
@merrymercy
Fix sliding window attention and gemma-2 unit tests in CI (sgl-projec…
0061128 
…t#1746)
@hnyls2002
Llama3.2 vision model support (sgl-project#1551)
94cde10
@hnyls2002
Update max_req_len and max_req_input_len (sgl-project#1748)
5e1558f
@merrymercy
Release v0.3.4.post1 (sgl-project#1749)
1f26e8b
@ByronHsu
Fix edge case for truncated (sgl-project#1747)
17536e7
@merrymercy
Fuse more ops & Simplify token mapping (sgl-project#1758)
ad4125d
@Ying1123 @ByronHsu
[API] add get memory pool size (sgl-project#1760)
2fce449 
Co-authored-by: Byron Hsu <[email protected]>
@merrymercy
Fix perf regression for set_kv_buffer (sgl-project#1765)
fbcbb26
@merrymercy
[Fix] Fix abort in dp (sgl-project#1767)
9af7b88
@merrymercy
Fix stop condition for <|eom_id|> (sgl-project#1766)
80a9054
@merrymercy @zhaochen20 @zhaochenyang20
Update docs (sgl-project#1768)
b7d0559 
Co-authored-by: Chayenne Zhao <[email protected]>
Co-authored-by: Chayenne <[email protected]>
@81549361
update
9441d92
@81549361 81549361 merged commit 197a9f6 into main Oct 23, 2024
0 of 3 checks passed
@81549361 81549361 deleted the sgl-project-main branch October 23, 2024 19:18
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.