Commits · 4f838c09cd7cdc8fd1b44b3d9ed9762c0e1e23e1 · zhaoyu6 / sglang

"benchmarks/communication/broadcast.py" did not exist on "1b2721adcd96656bb1f27d1f2f60947567b2d505"

19 Jun, 2025 7 commits
- [PD] Transfer hidden states for mtp when disaggregation (#7242) · 4f838c09
  Atream authored Jun 20, 2025
  
  4f838c09
- Put `_normalize_rid` before other normalization in `io_struct` (#7363) · d20a073b
  Chang Su authored Jun 19, 2025
  
  d20a073b
- [Refactor] Clean up radix cache related API (#7303) · 47367b76
  DarkSharpness authored Jun 19, 2025
```
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
```
  47367b76
- Reintroduce tiny fix sampler error when prob is not contiguous (#7354) · 650127a1
  fzyzcjy authored Jun 20, 2025
  
  650127a1
- Multi-Stage Awake: Support Resume and Pause KV Cache and Weights separately (#7099) · 3774f078
  Stefan He authored Jun 19, 2025
  
  3774f078
- add seed in CPU UTs to avoid flaky failure (#7333) · 9179ea15
  Chunyuan WU authored Jun 19, 2025
  
  9179ea15
- fix: resolve blackwell deepep image issue (#7331) · 20a503c7
  Yineng Zhang authored Jun 18, 2025
  
  20a503c7
18 Jun, 2025 12 commits
- Add more refactored openai test & in CI (#7284) · ffd1a26e
  Jinn authored Jun 18, 2025
  
  ffd1a26e
- Merge PDLB (Prefill-Decode Load Balancer) into SGLang Router (#7096) · 09ae5b20
  Simo Lin authored Jun 18, 2025
  
  09ae5b20
- [pd] optimize dockerfile for pd disaggregation (#7319) · 712bf9ec
  ybyang authored Jun 19, 2025
```
Co-authored-by: zhyncs <me@zhyncs.com>
```
  712bf9ec
- Fix profiler error when there are idle passes (#7003) · 9c6a0656
  fzyzcjy authored Jun 19, 2025
  
  9c6a0656
- Fix mini_lb for PD with long output: limit chunk size of decode response (#7301) · 2ae809c5
  ch-tiger1 authored Jun 19, 2025
```
Signed-off-by: ch-tiger1 <xyz@ch-tech.ip-ddns.com>
Co-authored-by: ch-tiger1 <xyz@ch-tech.ip-ddns.com>
```
  2ae809c5
- update invalid link in doc (#7297) · 1de4db9b
  linzhuo authored Jun 18, 2025
  
  1de4db9b
- chore: change logs from`INFO` to `DEBUG` for dp and add force quit for tokenizer manager (#7251) · 31fccf5a
  ishandhanani authored Jun 18, 2025
  
  31fccf5a
- Fix hicache benchmark script bug - some sampled input_request is [] (#7300) · b783c1cb
  Binyao Jiang authored Jun 17, 2025
  
  b783c1cb
- Update python API of activation, topk, norm and rope and remove vllm dependency (#6614) · 094c116f
  YanbingJiang authored Jun 18, 2025
```
Co-authored-by: Wu, Chunyuan <chunyuan.wu@intel.com>
Co-authored-by: jianan-gu <jianan.gu@intel.com>
Co-authored-by: sdp <sdp@gnr799219.jf.intel.com>
```
  094c116f
- Upstreaming hicache bug fixes (#7267) · e56685ac
  Zhiqiang Xie authored Jun 17, 2025
  
  e56685ac
- [PD] Add custom memory pool option to support Mooncake PD with NVLink (#7264) · c26d7349
  shangmingc authored Jun 18, 2025
```
Signed-off-by: Shangming Cai <caishangming@linux.alibaba.com>
```
  c26d7349
- [PD] Support get local ip from NIC for PD disaggregation (#7237) · ceaa85c9
  shangmingc authored Jun 18, 2025
```
Signed-off-by: Shangming Cai <caishangming@linux.alibaba.com>
```
  ceaa85c9
17 Jun, 2025 13 commits
- fix: only enable flash_attn test on sm80 sm90 (#7289) · 0650e517
  Yineng Zhang authored Jun 17, 2025
  
  0650e517
- ci: Fix test_ebnf_generate_all_optional_function_params (#7288) · fc554105
  Chang Su authored Jun 17, 2025
  
  fc554105
- fix: resolve b200 dsv3 mtp issue (#7286) · 4f204db5
  Yineng Zhang authored Jun 17, 2025
  
  4f204db5
- Fix AWQ Dequant and Weight Loading of deepseek v2 (#6842) · 3eb4a800
  AniZpZ authored Jun 18, 2025
  
  3eb4a800
- bugfix(tool call ebnf): Fix EBNF generation for optional function parameters (#7283) · e7261315
  Chang Su authored Jun 17, 2025
  
  e7261315
- Fix Deepseek R1 0528 FP4 tensor name mismatch issue during weights loading. (#7164) · 8c16da33
  Charles Chen authored Jun 17, 2025
  
  8c16da33
- support qwen2 running on ascend npu device (#7022) · a39d9287
  Yijie Zhu authored Jun 18, 2025
```
Co-authored-by: 刁莹煜 <diaoyingyu1@hisilicon.com>
```
  a39d9287
- feat: mtp support dp-attention (#6081) · 10d60cd4
  u4lr451 authored Jun 17, 2025
```
Co-authored-by: austindeng <austindeng@tencent.com>
Co-authored-by: tianqilin.99 <tianqilin.99@bytedance.com>
Co-authored-by: Qiaolin Yu <liin1211@outlook.com>
Co-authored-by: ch-wan <cwan39@gatech.edu>
```
  10d60cd4
- update ci node for xeon (#7265) · 8a10c4c3
  DiweiSun authored Jun 17, 2025
  
  8a10c4c3
- [amd] Opt dsv3 moe (#7160) · 405780bc
  kk authored Jun 17, 2025
```
Co-authored-by: wunhuang <wunhuang@amd.com>
```
  405780bc
- OAI Server Skeleton & Core Utility Endpoints (#7179) · 1dffee31
  yhyang201 authored Jun 17, 2025
  
  1dffee31
- [Refactor] OAI Server components (#7167) · 70c471a8
  Xinyuan Tong authored Jun 16, 2025
```
Signed-off-by: Xinyuan Tong <justinning0323@outlook.com>
```
  70c471a8
- Fix AMD speculative decoding (#7252) · 1a9c2c92
  Lianmin Zheng authored Jun 16, 2025
  
  1a9c2c92
16 Jun, 2025 8 commits
- support custom weight loader for model runner (#7122) · 873ae12c
  KavioYu authored Jun 17, 2025
```
Co-authored-by: kavioyu <kavioyu@tencent.com>
```
  873ae12c
- Use seq_len_fill_value in the cuda graph runners (#7233) · c64290dc
  Lianmin Zheng authored Jun 16, 2025
  
  c64290dc
- fix amd EP MoE FP8 issue (#7125) · 8e2363dc
  Alex Sun authored Jun 17, 2025
  
  8e2363dc
- chore: bump v0.4.7.post1 (#7248) · f9dc9dd2
  Yineng Zhang authored Jun 16, 2025
  
  f9dc9dd2
- Update CI flakes. (#7244) · 62a7aa2e
  Sai Enduri authored Jun 16, 2025
  
  62a7aa2e
- [fix] fix DeepGEMM blackwell input quant & ut & fix style and log (#7247) · 5ca07eed
  JieXin Liang authored Jun 17, 2025
  
  5ca07eed
- Feat/support rerank (#6058) · e30ef368
  woodx authored Jun 17, 2025
  
  e30ef368
- Tiny remove comments about DeepEP on H20 (#7234) · 91a066ec
  fzyzcjy authored Jun 17, 2025
  
  91a066ec