Commits · f9b567dfdc16f97850a73bb015bb64707a1143bc · OpenDAS / vllm_cscc

02 Jan, 2025 1 commit
- add softcap interface for gemma-2 · f9b567df
  zhuwenwen authored Jan 02, 2025
  
  f9b567df
13 Dec, 2024 1 commit
- remove ck fa interface · 54ddee7f
  zhuwenwen authored Dec 13, 2024
  
  54ddee7f
11 Dec, 2024 1 commit
- [feat]优化medusa代码，通过VLLM_TREE_DECODING环境变量控制是否采用tree-style解码，计算逻辑主干隔离 · 44e3ca68
  王敏 authored Dec 11, 2024
  
  44e3ca68
05 Dec, 2024 1 commit
- [Doc] Create a new "Usage" section (#10827) · aa39a8e1
  Cyrus Leung authored Dec 05, 2024
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  aa39a8e1
02 Dec, 2024 1 commit
- [misc] use out argument for flash attention (#10822) · a4c4daf3
  youkaichao authored Dec 02, 2024
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
```
  a4c4daf3
27 Nov, 2024 1 commit
- 1.优化medusa推理，节省cpu耗时 · 28375803
  王敏 authored Nov 27, 2024
```
2.更新medusa readme
3.解决benchmark_moe报错问题
```
  28375803
22 Nov, 2024 1 commit
- [torch.compile] support all attention backends (#10558) · eebad39f
  youkaichao authored Nov 22, 2024
```
Signed-off-by: youkaichao <youkaichao@gmail.com>
```
  eebad39f
07 Nov, 2024 1 commit
- [Feature] [Spec decode]: Combine chunked prefill with speculative decoding (#9291) · 9d43afcc
  Nicolò Lucchesi authored Nov 07, 2024
```
Signed-off-by: NickLucche <nlucches@redhat.com>
```
  9d43afcc
01 Nov, 2024 1 commit
- [Core][VLM] Add precise multi-modal placeholder tracking (#8346) · 6c0b7f54
  Peter Salas authored Nov 01, 2024
```
Signed-off-by: Peter Salas <peter@fixie.ai>
```
  6c0b7f54
26 Oct, 2024 1 commit
- Fix: MI100 Support By Bypassing Custom Paged Attention (#9560) · 55137e8e
  ErkinSagiroglu authored Oct 26, 2024
  
  55137e8e
24 Oct, 2024 1 commit
- 增加medusa并行解码功能,后续增加使用说明和测试文档 · 19bc93d9
  王敏 authored Oct 24, 2024
  
  19bc93d9
21 Oct, 2024 1 commit
- [Doc] Consistent naming of attention backends (#9498) · 496e991d
  Thomas Parnell authored Oct 21, 2024
```
Signed-off-by: Thomas Parnell <tpa@zurich.ibm.com>
```
  496e991d
12 Oct, 2024 1 commit
- [SpecDec] Remove Batch Expansion (2/3) (#9298) · 89feb4c8
  Lily Liu authored Oct 11, 2024
  
  89feb4c8
11 Oct, 2024 1 commit
- [Doc] Compatibility matrix for mutual exclusive features (#8512) · 8baf85e4
  Wallas Henrique authored Oct 11, 2024
```
Signed-off-by: Wallas Santos <wallashss@ibm.com>
```
  8baf85e4
06 Oct, 2024 1 commit
- [Bugfix] Fix incorrect updates to num_computed_tokens in multi-step scheduling (#9038) · cb3b2b9b
  Varun Sundar Rabindranath authored Oct 06, 2024
```
Co-authored-by: Varun Sundar Rabindranath <varun@neuralmagic.com>
```
  cb3b2b9b
01 Oct, 2024 1 commit
- [Spec Decode] (1/2) Remove batch expansion (#8839) · 15702038
  Lily Liu authored Oct 01, 2024
  
  15702038
27 Sep, 2024 2 commits
- [torch.compile] use empty tensor instead of None for profiling (#8875) · a9b15c60
  youkaichao authored Sep 27, 2024
  
  a9b15c60
- update version and deps · 5a9c236d
  zhuwenwen authored Sep 27, 2024
  
  5a9c236d
20 Sep, 2024 2 commits
- choose ck or cutlass implementation based on the fa version · 7fafc730
  zhuwenwen authored Sep 20, 2024
  
  7fafc730
- [bugfix] [AMD] add multi-step advance_step to ROCmFlashAttentionMetadata (#8474) · 9e5ec35b
  William Lin authored Sep 19, 2024
  
  9e5ec35b
19 Sep, 2024 1 commit
- [Kernel][Amd] Add fp8 kv cache support for rocm custom paged attention (#8577) · 9cc373f3
  Charlie Fu authored Sep 19, 2024
  
  9cc373f3
18 Sep, 2024 1 commit
- [CI/Build] Avoid CUDA initialization (#8534) · 6ffa3f31
  Cyrus Leung authored Sep 18, 2024
  
  6ffa3f31
14 Sep, 2024 1 commit
- [Kernel][Hardware][Amd]Custom paged attention kernel for rocm (#8310) · 1ef0d2ef
  Charlie Fu authored Sep 13, 2024
  
  1ef0d2ef
13 Sep, 2024 1 commit
- remove the automatic switching strategy of fa · 408f663a
  zhuwenwen authored Sep 13, 2024
  
  408f663a
10 Sep, 2024 1 commit
- update fa interface param · 5a9db327
  zhuwenwen authored Sep 10, 2024
  
  5a9db327
09 Sep, 2024 1 commit
- add cutlass fa and support bloom nn layout · 82f1ffdf
  zhuwenwen authored Sep 09, 2024
  
  82f1ffdf
20 Aug, 2024 1 commit
- [Core] Add `AttentionState` abstraction (#7663) · 3b682179
  Antoni Baum authored Aug 20, 2024
  
  3b682179
19 Aug, 2024 1 commit
- Add fa pad conditions and automatic switching strategy · b40f2ffc
  zhuwenwen authored Aug 19, 2024
  
  b40f2ffc
12 Aug, 2024 1 commit
- [Core/Bugfix] Add FP8 K/V Scale and dtype conversion for prefix/prefill Triton Kernel (#7208) · a046f863
  jon-chuang authored Aug 12, 2024
```
Co-authored-by: Cody Yu <hao.yu.cody@gmail.com>
```
  a046f863
10 Aug, 2024 1 commit
- add benchmarks to vllm and add env of VLLM_USE_FLASH_ATTN_AUTO · efb2f75f
  zhuwenwen authored Aug 10, 2024
  
  efb2f75f
01 Aug, 2024 1 commit
- [Misc] Support attention logits soft-capping with flash-attn (#7022) · 805a8a75
  Woosuk Kwon authored Aug 01, 2024
  
  805a8a75
25 Jul, 2024 1 commit
- skip window_size and alibi_slopes because ck fa is not supported, and add triton fa · 8e56ae5e
  zhuwenwen authored Jul 25, 2024
  
  8e56ae5e
20 Jul, 2024 1 commit
- [Bugfix][CI/Build][Hardware][AMD] Fix AMD tests, add HF cache, update CK FA,... · 06d6c5fe
  Matt Wong authored Jul 20, 2024
```
[Bugfix][CI/Build][Hardware][AMD] Fix AMD tests, add HF cache, update CK FA, add partially supported model notes (#6543)
```
  06d6c5fe
17 Jul, 2024 1 commit
- [Core] Refactor _prepare_model_input_tensors - take 2 (#6164) · 2fa4623d
  Cody Yu authored Jul 17, 2024
  
  2fa4623d
16 Jul, 2024 1 commit
- [Kernel][Attention] Separate `Attention.kv_scale` into `k_scale` and `v_scale` (#6081) · 978aed53
  Michael Goin authored Jul 16, 2024
  
  978aed53
08 Jul, 2024 1 commit

[Kernel] Correctly invoke prefill & decode kernels for cross-attention... · 543aa485

afeldman-nm authored Jul 08, 2024


[Kernel] Correctly invoke prefill & decode kernels for cross-attention (towards eventual encoder/decoder model support) (#4888)
Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>

543aa485

06 Jul, 2024 1 commit
- Use triton fa by default · 1863c926
  zhuwenwen authored Jul 06, 2024
  
  1863c926
04 Jul, 2024 1 commit

[ROCm][AMD][Model]Adding alibi slopes support in ROCm triton flash attention... · 56b325e9

Gregory Shtrasberg authored Jul 04, 2024


[ROCm][AMD][Model]Adding alibi slopes support in ROCm triton flash attention and naive flash attention (#6043)
Co-authored-by: Hongxia Yang <62075498+hongxiayang@users.noreply.github.com>

56b325e9

26 Jun, 2024 1 commit

[Core] Refactor Worker and ModelRunner to consolidate control plane communication (#5408) · dda48115

Stephanie Wang authored Jun 25, 2024


Signed-off-by: Stephanie Wang <swang@cs.berkeley.edu>
Signed-off-by: Stephanie <swang@anyscale.com>
Co-authored-by: Stephanie <swang@anyscale.com>

dda48115

08 Jun, 2024 1 commit
- [ROCm][AMD] Use pytorch sdpa math backend to do naive attention (#4965) · c96fc067
  Hongxia Yang authored Jun 07, 2024
  
  c96fc067