Commit Graph - vllm - Gitea: Git with a cup of tea

wassname/vllm

Fork 0

mirror of https://github.com/wassname/vllm.git synced 2026-09-10 12:41:29 +08:00

4dd6416faf Fix stablelm (#3038) Roy 2024-02-27 10:31:10 +08:00
c1c0d00b88 Don't use cupy when enforce_eager=True (#3037) Roy 2024-02-27 09:33:38 +08:00
d9f726c4d0 [Minor] Remove unused config files (#3039) Roy 2024-02-27 09:25:22 +08:00
d6e4a130b0 [Minor] Remove gather_cached_kv kernel (#3043) Woosuk Kwon 2024-02-26 15:00:54 -08:00
cfc15a1031 Optimize Triton MoE Kernel (#2979) Philipp MoritzandCade Daniel 2024-02-26 13:48:56 -08:00
70f3e8e3a1 Add LogProbs for Chat Completions in OpenAI (#2918) Jared Moore 2024-02-25 18:39:34 -08:00
ef978fe411 Port metrics from aioprometheus to prometheus_client (#2730) Harry Mellor 2024-02-25 19:54:00 +00:00
f7c1234990 [Fix] Fissertion on YaRN model len (#2984) Woosuk Kwon 2024-02-23 12:57:48 -08:00
57f044945f Fix nvcc not found in vlm-openai image (#2781) zhaoyang-star 2024-02-23 06:25:07 +08:00
4caf7044e0 Include tokens from prompt phase in counter_generation_tokens (#2802) Ronen Schaffer 2024-02-23 00:00:12 +02:00
6f32cddf1c Remove Flash Attention in test env (#2982) Woosuk Kwon 2024-02-22 09:58:29 -08:00
c530e2cfe3 [FIX] Fix a bug in initializing Yarn RoPE (#2983) 44670 2024-02-22 17:40:05 +08:00
fd5dcc5c81 Optimize GeGLU layer in Gemma (#2975) Woosuk Kwon 2024-02-21 20:17:52 -08:00
93dc5a2870 chore(vllm): codespell for spell checking (#2820) Massimiliano Pronesti 2024-02-22 02:56:01 +00:00
95529e3253 Use Llama RMSNorm custom op for Gemma (#2974) Woosuk Kwon 2024-02-21 18:28:23 -08:00
344020c926 Migrate MistralForCausalLM to LlamaForCausalLM (#2868) Roy 2024-02-22 10:25:05 +08:00
5574081c49 Added early stopping to completion APIs (#2939) Mustafa Eyceoz 2024-02-21 21:24:01 -05:00
d7f396486e Update comment (#2934) Ronen Schaffer 2024-02-22 04:18:37 +02:00
8fbd84bf78 Bump up version to v0.3.2 (#2968) Zhuohan Li 2024-02-21 11:47:25 -08:00
7d2dcce175 Support per-request seed (#2514) Nick Hill 2024-02-21 11:47:00 -08:00
dc903e70ac [ROCm] Upgrade transformers to v4.38.0 (#2967) Woosuk Kwon 2024-02-21 09:46:57 -08:00
a9c8212895 [FIX] Add Gemma model to the doc (#2966) Zhuohan Li 2024-02-21 09:46:15 -08:00
c20ecb6a51 Upgrade transformers to v4.38.0 (#2965) Woosuk Kwon 2024-02-21 09:38:03 -08:00
5253edaacb Add Gemma model (#2964) Xiang Xu 2024-02-21 09:34:30 -08:00
017d9f1515 Add metrics to RequestOutput (#2876) Antoni Baum 2024-02-20 21:55:57 -08:00
181b27d881 Make vLLM logging formatting optional (#2877) Antoni Baum 2024-02-20 14:38:55 -08:00
63e2a6419d [FIX] Fix beam search test (#2930) Zhuohan Li 2024-02-20 14:37:39 -08:00
264017a2bf [ROCm] include gfx908 as supported (#2792) James Whedbee 2024-02-19 19:58:59 -06:00
e433c115bc Fix vllm:prompt_tokens_total metric calculation (#2869) Ronen Schaffer 2024-02-19 09:55:41 +02:00
86fd8bb0ac Add warning to prevent changes to benchmark api server (#2858) Simon Mo 2024-02-18 21:36:19 -08:00
ab3a5a8259 Support OLMo models. (#2832) Isotr0py 2024-02-19 13:05:15 +08:00
a61f0521b8 [Test] Add basic correctness test (#2908) Zhuohan Li 2024-02-18 16:44:50 -08:00
537c9755a7 [Minor] Small fix to make distributed init logic in worker looks cleaner (#2905) Zhuohan Li 2024-02-18 14:39:00 -08:00
786b7f18a5 Add code-revision config argument for Hugging Face Hub (#2892) Mark Mozolewski 2024-02-17 22:36:53 -08:00
8f36444c4f multi-LoRA as extra models in OpenAI server (#2775) jvmncs 2024-02-17 15:00:48 -05:00
185b2c29e2 Defensively copy sampling_params (#2881) Nick Hill 2024-02-17 11:18:04 -08:00
5f08050d8d Bump up to v0.3.1 (#2887) Woosuk Kwon 2024-02-16 15:05:18 -08:00
64da65b322 Prefix Caching- fix t4 triton error (#2517) shiyi.c_98 2024-02-16 14:17:55 -08:00
5255d99dc5 [ROCm] Dockerfile fix for flash-attention build (#2885) Hongxia Yang 2024-02-15 13:22:39 -05:00
4f2ad11135 Fix DeciLM (#2883) Philipp Moritz 2024-02-14 22:29:57 -08:00
d7afab6d3a [BugFix] Fix GC bug for LLM class (#2882) Woosuk Kwon 2024-02-14 22:17:44 -08:00
31348dff03 Align LoRA code between Mistral and Mixtral (fixes #2875) (#2880) Philipp MoritzandPernekhan Utemuratov 2024-02-14 16:00:43 -08:00
25e86b6a61 Don't use cupy NCCL for AMD backends (#2855) Woosuk Kwon 2024-02-14 12:30:44 -08:00
4efbac6d35 Migrate AquilaForCausalLM to LlamaForCausalLM (#2867) Roy 2024-02-15 04:30:24 +08:00
87069ccf68 Fix docker python version (#2845) Nikola Borisov 2024-02-14 10:17:57 -08:00
7e45107f51 [Fix] Fix memory profiling when GPU is used by multiple processes (#2863) Woosuk Kwon 2024-02-13 19:52:34 -08:00
0c48b37c31 Fix internlm after https://github.com/vllm-project/vllm/pull/2860 (#2861) Philipp Moritz 2024-02-13 18:01:15 -08:00
7eacffd951 Migrate InternLMForCausalLM to LlamaForCausalLM (#2860) Philipp MoritzandRoy 2024-02-13 17:12:05 -08:00
2a543d6efe Add LoRA support for Mixtral (#2831) Terry 2024-02-13 15:55:45 -08:00
317b29de0f Remove Yi model definition, please use LlamaForCausalLM instead (#2854) Philipp MoritzandRoy 2024-02-13 14:22:22 -08:00
a463c333dd Use CuPy for CUDA graphs (#2811) Woosuk Kwon 2024-02-13 11:32:06 -08:00
ea356004d4 Revert "Refactor llama family models (#2637)" (#2851) Philipp Moritz 2024-02-13 09:24:59 -08:00
5c976a7e1a Refactor llama family models (#2637) Roy 2024-02-13 16:09:23 +08:00
f964493274 [CI] Ensure documentation build is checked in CI (#2842) Simon Mo 2024-02-12 22:53:07 -08:00
a4211a4dc3 Serving Benchmark Refactoring (#2433) Roger Wang 2024-02-12 22:53:00 -08:00
563836496a Refactor 2 awq gemm kernels into m16nXk32 (#2723) RexandChunan Zeng 2024-02-12 11:02:17 -08:00
4ca2c358b1 Add documentation section about LoRA (#2834) Philipp Moritz 2024-02-12 08:24:45 -08:00
0580aab02f [ROCm] support Radeon™ 7900 series (gfx1100) without using flash-attention (#2768) Hongxia Yang 2024-02-11 02:14:37 -05:00
3711811b1d Disable custom all reduce by default (#2808) Woosuk Kwon 2024-02-08 09:58:03 -08:00
65b89d16ee [Ray] Integration compiled DAG off by default (#2471) SangBin Cho 2024-02-09 02:57:25 +09:00
931746bc6d Add documentation on how to do incremental builds (#2796) Philipp Moritz 2024-02-07 14:42:02 -08:00
c81dddb45c [ROCm] Fix build problem resulted from previous commit related to FP8 kv-cache support (#2790) Hongxia Yang 2024-02-07 01:36:59 -05:00
fe6d09ae61 [Minor] More fix of test_cache.py CI test failure (#2750) Lily Liu 2024-02-06 11:38:38 -08:00
ed70c70ea3 modelscope: fix issue when model parameter is not a model id but path of the model. (#2489) liuyhwangyh 2024-02-07 01:57:15 +08:00
f0d4e14557 Add fused top-K softmax kernel for MoE (#2769) Woosuk Kwon 2024-02-05 17:38:02 -08:00
2ccee3def6 [ROCm] Fixup arch checks for ROCM (#2627) Douglas Lehr 2024-02-05 16:59:09 -06:00
b92adec8e8 Set local logging level via env variable (#2774) Lukas 2024-02-05 23:26:50 +01:00
56f738ae9b [ROCm] Fix some kernels failed unit tests (#2498) Hongxia Yang 2024-02-05 17:25:36 -05:00
72d3a30c63 [Minor] Fix benchmark_latency script (#2765) Woosuk Kwon 2024-02-05 12:45:37 -08:00
c9b45adeeb Require triton >= 2.1.0 (#2746) whyiugandyangrui1 2024-02-05 15:07:36 +08:00
5a6c81b051 Remove eos tokens from output by default (#2611) Rex 2024-02-04 14:32:42 -08:00
51cd22ce56 set&get llm internal tokenizer instead of the TokenizerGroup (#2741) dancingpipiandshujunhua1 2024-02-05 06:25:36 +08:00
5ed704ec8c docs: fix langchain (#2736) Massimiliano Pronesti 2024-02-04 03:17:55 +01:00
4abf6336ec Add one example to run batch inference distributed on Ray (#2696) Cheng Su 2024-02-02 15:41:42 -08:00
0e163fce18 Fix default length_penalty to 1.0 (#2667) zspo 2024-02-02 07:59:39 +08:00
96b6f475dd Remove hardcoded device="cuda" to support more devices (#2503) Kunshang JiandJiang Li 2024-02-02 07:46:39 +08:00
c410f5d020 Use revision when downloading the quantization config file (#2697) Pernekhan UtemuratovandPernekhan Utemuratov 2024-02-01 15:41:58 -08:00
bb8c697ee0 Update README for meetup slides (#2718) Simon Mo 2024-02-01 14:56:53 -08:00
b9e96b17de fix python 3.8 syntax (#2716) Simon Mo 2024-02-01 14:00:58 -08:00
923797fea4 Fix compile error when using rocm (#2648) zhaoyang-star 2024-02-02 01:35:09 +08:00
cd9e60c76c Add Internlm2 (#2666) Fengzhe Zhou 2024-02-02 01:27:40 +08:00
93b38bea5d Refactor Prometheus and Add Request Level Metrics (#2316) Robert Shaw 2024-01-31 14:58:07 -08:00
d0d93b92b1 Add unit test for Mixtral MoE layer (#2677) Philipp Moritz 2024-01-31 14:34:17 -08:00
89efcf1ce5 [Minor] Fix test_cache.py CI test failure (#2684) Philipp Moritz 2024-01-31 10:12:11 -08:00
c664b0e683 fix some bugs (#2689) zspo 2024-02-01 02:09:23 +08:00
d69ff0cbbb Fixes assertion failure in prefix caching: the lora index mapping should respect prefix_len (#2688) Tao He 2024-02-01 01:00:13 +08:00
1af090b57d Bump up version to v0.3.0 (#2656) Zhuohan Li 2024-01-31 00:07:07 -08:00
3dad944485 Add quantized mixtral support (#2673) Woosuk Kwon 2024-01-30 16:34:10 -08:00
105a40f53a [Minor] Fix false warning when TP=1 (#2674) Woosuk Kwon 2024-01-30 14:39:40 -08:00
bbe9bd9684 [Minor] Fix a small typo (#2672) Philipp Moritz 2024-01-30 13:40:37 -08:00
4f65af0e25 Add swap_blocks unit tests (#2616) Vladimir 2024-01-30 18:30:50 +01:00
d79ced3292 Fix 'Actor methods cannot be called directly' when using --engine-use-ray (#2664) Wen Sun 2024-01-31 00:17:05 +08:00
ab40644669 Fused MOE for Mixtral (#2542) Philipp Moritzandchen shen 2024-01-29 22:43:37 -08:00
5d60def02c DeepseekMoE support with Fused MoE kernel (#2453) wangding zengandroy 2024-01-30 13:19:48 +08:00
ea8489fce2 ROCm: Allow setting compilation target (#2581) Rasmus Larsen 2024-01-29 19:52:31 +01:00
1b20639a43 No repeated IPC open (#2642) Hanzhi Zhou 2024-01-30 02:46:29 +08:00
b72af8f1ed Fix error when tp > 1 (#2644) zhaoyang-starandzhaoyang-star 2024-01-29 14:47:39 +08:00
9090bf02e7 Support FP8-E5M2 KV Cache (#2279)
zhaoyang-star zhaoyang Zhuohan Li
2024-01-29 08:43:54 +08:00
7d648418b8 Update Ray version requirements (#2636) Simon Mo 2024-01-28 14:27:22 -08:00
89be30fa7d Small async_llm_engine refactor (#2618) Murali Andoorveedu 2024-01-27 23:28:37 -08:00

... 41 42 43 44 45 ...

Commit Graph Select branches Hide Pull Requests main Mono Color

Commit Graph

Select branches

Hide Pull Requests

main