Commits · 1fe1a4b6588dee4cc43df0f688c5d62f4fd422f3 · OpenDAS / vllm_cscc

06 Dec, 2024 1 commit
- [feat]并行解码支持多卡推理 · 4a4e3601
  王敏 authored Dec 06, 2024
  
  4a4e3601
05 Dec, 2024 4 commits
- skip tests about cross-attn and not supported · 78800ecf
  zhuwenwen authored Dec 05, 2024
  
  78800ecf
- update DCU info and skip tests same to nv · 129fce94
  zhuwenwen authored Dec 05, 2024
  
  129fce94
- [fix]修复单测test_mlp_correctness失败问题 · d623e721
  王敏 authored Dec 05, 2024
  
  d623e721
- change AMD GPU and dcu to hcu · 3f78216a
  zhuwenwen authored Dec 05, 2024
  
  3f78216a
04 Dec, 2024 3 commits
- remove unsupported tests from kernels and add pytest html · a3d96521
  zhuwenwen authored Dec 04, 2024
  
  a3d96521
- 修复新request kv cache移动错误 · f0021a4d
  王敏 authored Dec 04, 2024
  
  f0021a4d
- support falcon and optimize layout · 9ddee6b1
  zhuwenwen authored Dec 04, 2024
  
  9ddee6b1
03 Dec, 2024 5 commits
- 修改cutlass 单测 · 26513bb5
  gaoqiong authored Dec 03, 2024
  
  26513bb5
- add large_gpu_test · 0dc55ec0
  zhuwenwen authored Dec 03, 2024
  
  0dc55ec0
- 增加w8a8的triton调度支持 · 1a9775b8
  gaoqiong authored Dec 03, 2024
  
  1a9775b8
- 增加w8a8的triton调度支持 · 1c77f16e
  gaoqiong authored Dec 03, 2024
  
  1c77f16e
- [fix]修复llm_engine.py 越界报错 · acdbc978
  王敏 authored Dec 03, 2024
  
  acdbc978
02 Dec, 2024 2 commits
- add ChatGLMForConditionalGeneration · 6ebda263
  zhuwenwen authored Dec 02, 2024
  
  6ebda263
- update qwen2 supported_lora_modules · d5723fbb
  zhuwenwen authored Dec 02, 2024
  
  d5723fbb
29 Nov, 2024 1 commit
- add VLLM_OPTEST_URLS_PORT to load https from local · 41b09879
  zhuwenwen authored Nov 29, 2024
  
  41b09879
28 Nov, 2024 1 commit
- update lm_head layout of chatglm · 77ae0f0d
  zhuwenwen authored Nov 28, 2024
  
  77ae0f0d
27 Nov, 2024 3 commits
- update tests · 87a2e37f
  zhuwenwen authored Nov 27, 2024
  
  87a2e37f
- 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
- add VLLM_OPTEST_MODELS_PATH/OPTEST_MODELS_PATH to load models from local path... · 3c9817d2
  zhuwenwen authored Nov 27, 2024
```
add VLLM_OPTEST_MODELS_PATH/OPTEST_MODELS_PATH  to load models from local path instead of Hugging Face Hub
```
  3c9817d2
25 Nov, 2024 1 commit
- update the lm-head layout of llama and qwen series models · 122d7146
  zhuwenwen authored Nov 25, 2024
  
  122d7146
20 Nov, 2024 3 commits
- [fix]修复单测test_mlp_correctness.py运行时的崩溃问题 · 96dcaff9
  王敏 authored Nov 20, 2024
  
  96dcaff9
- [fix]修复单测test_medusa_correctness.py中的错误 · 215f33b0
  王敏 authored Nov 20, 2024
  
  215f33b0
- [fix]修复单测test_encoder_decoder_model_runner报错问题 · 93089fb2
  王敏 authored Nov 20, 2024
  
  93089fb2
19 Nov, 2024 1 commit
- [fix]解决单测test_encode报错问题 · 4e23f2ad
  王敏 authored Nov 19, 2024
  
  4e23f2ad
18 Nov, 2024 2 commits
- [fix]解决test_model_input中报的找不到move_cache错误 · 4d29e0a8
  王敏 authored Nov 18, 2024
  
  4d29e0a8
- [fix]修复test_spec_decode_worker中的错误 · dc2ce261
  王敏 authored Nov 18, 2024
  
  dc2ce261
15 Nov, 2024 1 commit
- [fix]修复test_long_context中报错问题，单测依然无法通过，nv也是同样的问题 · 1d6cfb11
  王敏 authored Nov 15, 2024
  
  1d6cfb11
12 Nov, 2024 1 commit
- update idefics2_vision_model.py · 9a68d3a8
  zhuwenwen authored Nov 12, 2024
  
  9a68d3a8
07 Nov, 2024 2 commits
- remove xformers deps · 55c5f16f
  zhuwenwen authored Nov 07, 2024
  
  55c5f16f
- 1.Mixtral模型的gemm由TN改为NN · e7d022fd
  王敏 authored Nov 07, 2024
  
  e7d022fd
06 Nov, 2024 2 commits
- optimize qwen2-moe layout · 82aee745
  zhuwenwen authored Nov 06, 2024
  
  82aee745
- 1.解决medusa模型多卡推理结果异常问题 · f9060e6b
  王敏 authored Nov 06, 2024
```
2.examples中添加medusa readme
3.修复model_runner中input_positions配置错误的笔误，解决多个模型运行失败问题
```
  f9060e6b
04 Nov, 2024 1 commit
- optimize qwen2-vl layout · a4167889
  zhuwenwen authored Nov 04, 2024
  
  a4167889
01 Nov, 2024 1 commit
- support multi image input of local files · a89ac72c
  zhuwenwen authored Nov 01, 2024
  
  a89ac72c
30 Oct, 2024 2 commits
- remove unused code · 04d33670
  zhuwenwen authored Oct 30, 2024
  
  04d33670
- auto convert lm_head layout of llama · 1428c17d
  zhuwenwen authored Oct 30, 2024
  
  1428c17d
29 Oct, 2024 3 commits
- update llama.py · 85def94c
  zhuwenwen authored Oct 29, 2024
  
  85def94c
- [Model] Add telechat-12b and GLM-4v support · 1e4ba896
  zhuwenwen authored Oct 29, 2024
  
  1e4ba896
- [Model] Add GLM-4v support · cbdc3a13
  zhuwenwen authored Oct 29, 2024
  
  cbdc3a13