Commit Graph

Select branches

Hide Pull Requests

main

Mono Color

• eb8fcd2666 [BugFix] Apply get_cached_tokenizer to the tokenizer setter of LLM (#5207) 3 people 2024-06-06 01:59:02 +08:00
• 5563a4dea8 [Model] Correct Mixtral FP8 checkpoint loading (#5231) Cody YuandGitHub 2024-06-05 10:58:50 -07:00
• ccd4f129e8 [Kernel] Add GPU architecture guards to the CUTLASS w8a8 kernels to reduce binary size (#5157) 3 people 2024-06-05 13:44:15 -04:00
• 02cc3b51a7 [misc] benchmark_serving.py -- add ITL results and tweak TPOT results (#5263) Tyler Michael SmithandGitHub 2024-06-05 13:17:51 -04:00
• d5b1eb081e [CI] Add nightly benchmarks (#5260) Simon MoandGitHub 2024-06-05 11:42:08 -05:00
• f0a500545f [Frontend] OpenAI API server: Add add_special_tokens to ChatCompletionRequest (default False) (#5278) tomeras91andGitHub 2024-06-05 19:32:58 +03:00
• c65146e75e [Misc] Fix docstring of get_attn_backend (#5271) Woosuk KwonandGitHub 2024-06-05 09:18:59 -07:00
• 41ca62cf03 [Misc] Add CustomOp interface for device portability (#5255) Woosuk KwonandGitHub 2024-06-05 09:18:19 -07:00
• 974fc9b845 [Bugfix] Fix prompt_logprobs when SamplingParams.detokenize is set to True (#5226) zifeitongandGitHub 2024-06-04 19:37:28 -07:00
• fee4dcc33a [Misc] update collect env (#5261) youkaichaoandGitHub 2024-06-04 15:29:09 -07:00
• 650a4cc55e [Misc] Add transformers version to collect_env.py (#5259) Michael GoinandGitHub 2024-06-04 15:52:28 -04:00
• 9ca62d8668 [CI] mark AMD test as softfail to prevent blockage (#5256) Simon MoandGitHub 2024-06-04 13:34:53 -05:00
• 45c35f0d58 [CI/Build] Reducing CPU CI execution time (#5241) Li, JiangandGitHub 2024-06-05 01:26:40 +08:00
• 9ba093b4f4 [CI/Build] Simplify model loading for HfRunner (#5251) Cyrus LeungandGitHub 2024-06-05 01:09:19 +08:00
• 27208be66e [Kernel] Add back batch size 1536 and 3072 to MoE tuning (#5242) Woosuk KwonandGitHub 2024-06-04 09:58:47 -07:00
• 87d5abef75 [Bugfix] Fix a bug caused by pip install setuptools>=49.4.0 for CPU backend (#5249) Jie Fu (傅杰)andGitHub 2024-06-05 00:57:51 +08:00
• ec784b2526 [CI/Build] Add inputs tests (#5215) Cyrus LeungandGitHub 2024-06-04 12:01:46 +08:00
• a58f24e590 [Bugfix] Fix torch.compile() error when using MultiprocessingGPUExecutor (#5229) zifeitongandGitHub 2024-06-03 20:55:50 -07:00
• f42a006b15 [Bugfix]: During testing, use pytest monkeypatch for safely overriding the env var that indicates the vLLM backend (#5210) afeldman-nmandGitHub 2024-06-03 23:32:57 -04:00
• 3a434b07ed [Kernel] Enhance MoE benchmarking & tuning script (#4921) Woosuk KwonandGitHub 2024-06-03 20:06:59 -07:00
• bd0e7802e0 [Bugfix] Add warmup for prefix caching example (#5235) Zhuohan LiandGitHub 2024-06-03 19:36:41 -07:00
• 06b2550cbb [Bugfix] Support prompt_logprobs==0 (#5217) Toshiki KataokaandGitHub 2024-06-04 09:59:30 +09:00
• f775a07e30 [FRONTEND] OpenAI tools support named functions (#5032) Breno FariaandGitHub 2024-06-04 01:25:29 +02:00
• 4f0d17c05c New CI template on AWS stack (#5110) Kevin H. LuuandGitHub 2024-06-03 16:16:43 -07:00
• 10c38e3e46 [Misc]: Implement CPU/GPU swapping in BlockManagerV2 (#3834) Kaiyang ChenandGitHub 2024-06-04 04:37:11 +08:00
• cafb8e06c5 [CI/BUILD] enable intel queue for longer CPU tests (#4113) YuanandGitHub 2024-06-04 01:39:50 +08:00
• cbb2f59cc8 [Kernel] Pass a device pointer into the quantize kernel for the scales (#5159) Tyler Michael SmithandGitHub 2024-06-03 12:52:30 -04:00
• 0ab278ca31 [Core] Remove unnecessary copies in flash attn backend (#5138) Antoni BaumandGitHub 2024-06-03 09:39:31 -07:00
• 7a64d24aad [Core] Support image processor (#4197) Cyrus LeungandGitHub 2024-06-03 13:56:41 +08:00
• dfbe60dc62 [Misc] Simplify code and fix type annotations in conftest.py (#5118) Cyrus LeungandGitHub 2024-06-03 07:05:50 +08:00
• a66cf40b20 [Kernel][ROCm][AMD] enable fused topk_softmax kernel for moe layer (#4927) Divakar VermaandGitHub 2024-06-02 16:13:26 -05:00
• f790ad3c50 [Frontend][OpenAI] Support for returning max_model_len on /v1/models response (#4643) Avinash RajandGitHub 2024-06-02 13:36:13 +05:30
• ed59a7ed23 Update test_ignore_eos (#4898) Simon MoandGitHub 2024-06-01 21:21:53 -05:00
• 044793d8df [BugFix] Prevent LLM.encode for non-generation Models (#5184) 3 people 2024-06-01 19:35:41 -04:00
• c2d6d2f960 [Bugfix]: Fix issues related to prefix caching example (#5177) (#5180) Daniil ArapovandGitHub 2024-06-02 01:53:52 +03:00
• 8279078e21 [Bugfix] Remove deprecated @abstractproperty (#5174) Zhuohan LiandGitHub 2024-06-01 15:40:25 -07:00
• b9c0605a8e [Feature][Kernel] Support bitsandbytes quantization and QLoRA (#4776) chenqianfzhandGitHub 2024-06-01 13:51:10 -07:00
• 37464a0f74 [Bugfix] Fix call to init_logger in openai server (#4765) Nadav ShmayovitsandGitHub 2024-06-01 20:18:50 +03:00
• c354072828 [Minor] Fix the path typo in loader.py: save_sharded_states.py -> save_sharded_state.py (#5151) Ye CaoandGitHub 2024-06-02 01:11:22 +08:00
• f081c3ce4b [Kernel] Update Cutlass fp8 configs (#5144) 4 people 2024-06-01 14:16:07 +05:30
• 260d119e86 [Kernel] Refactor CUTLASS kernels to always take scales that reside on the GPU (#5137) Tyler Michael SmithandGitHub 2024-06-01 02:45:32 -04:00
• a360ff80bb [CI/Build] CMakeLists: build all extensions' cmake targets at the same time (#5034) DanieleandGitHub 2024-06-01 06:06:45 +02:00
• 1197e02141 [Build] Guard against older CUDA versions when building CUTLASS 3.x kernels (#5168) Tyler Michael SmithandGitHub 2024-05-31 20:21:38 -04:00
• 657579113f [Doc] Add checkmark for GPTBigCodeForCausalLM LoRA support (#5171) Nick HillandGitHub 2024-05-31 17:20:19 -07:00
• e9899fb7a4 [Model] Enable FP8 QKV in MoE and refine kernel tuning script (#5039) Cody YuandGitHub 2024-05-31 14:29:19 -07:00
• a377f0bd5e [Misc]: optimize eager mode host time (#4196) 3 people 2024-05-31 13:14:50 +08:00
• e9d3aa04f6 Revert "[Kernel] Marlin_24: Ensure the mma.sp instruction is using the ::ordered_metadata modifier (introduced with PTX 8.5)" (#5149) Simon MoandGitHub 2024-05-31 00:00:26 -05:00
• a22dea54d3 [Model] Support MAP-NEO model (#5081) 3 people 2024-05-31 10:24:41 +08:00
• 533c217792 Fix cutlass sm_90a vesrion in CMakeList simon-mo 2024-05-31 02:13:01 +00:00
• 6d21fa1cad [Kernel] Marlin_24: Ensure the mma.sp instruction is using the ::ordered_metadata modifier (introduced with PTX 8.5) (#5136) Alexander MatveevandGitHub 2024-05-30 22:02:11 -04:00
• b35be5403f [Bugfix] Avoid Warnings in SparseML Activation Quantization (#5120) Robert ShawandGitHub 2024-05-30 17:04:37 -07:00
• 45a1a69b98 [Build] Disable sm_90a in cu11 (#5141) Simon MoandGitHub 2024-05-30 16:37:16 -05:00
• 87a658c812 Bump version to v0.4.3 (#5046) Simon MoandGitHub 2024-05-30 13:13:46 -05:00
• 429d89720e add doc about serving option on dstack (#3074) 3 people 2024-05-31 02:11:07 +09:00
• a9bcc7afb2 [Doc] Use intersphinx and update entrypoints docs (#5125) Cyrus LeungandGitHub 2024-05-31 00:59:23 +08:00
• d79d9eaaff [Misc] remove duplicate definition of seq_lens_tensor in model_runner.py (#5129) Hyunsung LeeandGitHub 2024-05-30 22:56:19 +09:00
• f758505c73 [CI/Build] increase wheel size limit to 200 MB (#5130) youkaichaoandGitHub 2024-05-30 06:29:48 -07:00
• d910816c73 [Bugfix] Automatically Detect SparseML models (#5119) Robert ShawandGitHub 2024-05-30 05:58:37 -07:00
• 87d41c849d [BUGFIX] [FRONTEND] Correct chat logprobs (#5029) 3 people 2024-05-30 11:52:14 +02:00
• e07aff9e52 [CI/Build] Docker cleanup functionality for amd servers (#5112) 6 people 2024-05-29 22:27:39 -05:00
• 5bf185a1c4 [Bugfix] gptq_marlin: Ensure g_idx_sort_indices is not a Parameter (#5108) Alexander MatveevandGitHub 2024-05-29 20:30:18 -04:00
• 4fbcb0f27e [Doc][Build] update after removing vllm-nccl (#5103) 3 people 2024-05-29 16:51:18 -07:00
• 7c3604fb68 [Bugfix] logprobs is not compatible with the OpenAI spec #4795 (#5031) Itay EtelisandGitHub 2024-05-30 02:13:22 +03:00
• b1c255630d [Core] Avoid the need to pass None values to Sequence.inputs (#5099) Cyrus LeungandGitHub 2024-05-30 07:05:01 +08:00
• eb6c50cdc2 [Bugfix][CI/Build] Fix codespell failing to skip files in git diff (#5097) Cyrus LeungandGitHub 2024-05-30 07:02:54 +08:00
• eecd864388 [Bugfix][CI/Build] Fix test and improve code for merge_async_iterators (#5096) Cyrus LeungandGitHub 2024-05-30 07:02:25 +08:00
• ae495c74ea [Doc]Replace deprecated flag in readme (#4526) Ronen SchafferandGitHub 2024-05-30 01:26:33 +03:00
• 4238bc82f2 [Core] Cross-attention KV caching and memory-management (towards eventual encoder/decoder model support) (#4837) afeldman-nmandGitHub 2024-05-29 12:09:13 -04:00
• 594392d27a [Core][Distributed] improve p2p access check (#4992) youkaichaoandGitHub 2024-05-29 04:29:07 -07:00
• 18c1f16d86 [Bugfix] Fix arguments passed to Sequence in stop checker test (#5092) Cyrus LeungandGitHub 2024-05-29 15:16:41 +08:00
• 5bd3c65072 [Core][Optimization] remove vllm-nccl (#5091) youkaichaoandGitHub 2024-05-28 22:13:52 -07:00
• 616e600e0b [Misc] add gpu_memory_utilization arg (#5079) Marut PandyaandGitHub 2024-05-28 17:16:18 -07:00
• dfba529b40 [Bugfix] Remove the last EOS token unless explicitly specified (#5077) Junichi SatoandGitHub 2024-05-29 09:15:35 +09:00
• 5ae5ed1e60 [Core] Consolidate prompt arguments to LLM engines (#4328) 3 people 2024-05-29 04:29:31 +08:00
• 290f4ada2b [Docs] Add Dropbox as sponsors (#5089) Simon MoandGitHub 2024-05-28 12:29:09 -05:00
• dd8de11f0a [Kernel][ROCm][AMD] Add fused_moe Triton configs for MI300X (#4951) Divakar VermaandGitHub 2024-05-28 11:03:23 -05:00
• 9ba415588a [BugFix] Fix Embedding Models with TP>1 (#5075) Robert ShawandGitHub 2024-05-28 08:32:42 -07:00
• d4f3985907 [Core] Sliding window for block manager v2 (#4545) 3 people 2024-05-27 19:07:07 -07:00
• 890aa93d27 [Model] Add support for falcon-11B (#5069) Isotr0pyandGitHub 2024-05-28 07:41:43 +08:00
• fbdb7b3ee2 [Core] Allow AQLM on Pascal (#5058) sasha0552andGitHub 2024-05-27 22:26:14 +00:00
• 1102bef219 [Bugfix / Core] Prefix Caching Guards (merged with main) (#4846) 4 people 2024-05-27 15:18:17 -07:00
• f17a1a8f96 [Misc] Make Serving Benchmark More User-friendly (#5044) Roger WangandGitHub 2024-05-25 10:28:16 -07:00
• d5a1697772 [Dynamic Spec Decoding] Minor fix for disabling speculative decoding (#5000) Lily LiuandGitHub 2024-05-25 10:00:14 -07:00
• 325c119961 [Misc] add logging level env var (#5045) youkaichaoandGitHub 2024-05-24 23:49:49 -07:00
• 8e192ff967 [Kernel][Backend][Model] Blocksparse flash attention kernel and Phi-3-Small model (#4799) 6 people 2024-05-25 01:00:52 -04:00
• e64fde4b01 [Core][Bugfix]: fix prefix caching for blockv2 (#4764) 3 people 2024-05-25 01:07:09 +08:00
• 919770957f [Bugfix] Fix Mistral v0.3 Weight Loading (#5005) 3 people 2024-05-24 14:28:27 +02:00
• 6a50f4cafa [Doc] add ccache guide in doc (#5012) 3 people 2024-05-23 16:21:54 -07:00
• e3470f8753 [Core]: Option To Use Prompt Token Ids Inside Logits Processor (#4985) 3 people 2024-05-24 01:04:24 +03:00
• a1242324c9 [Kernel] Initial Activation Quantization Support (#4525) 4 people 2024-05-23 17:29:18 -04:00
• 5eda2ea02a [Core][1/N] Support send/recv in PyNCCL Groups (#4988) Murali AndoorveeduandGitHub 2024-05-23 09:54:48 -07:00
• 2ba80bed27 [Bugfix] Update Dockerfile.cpu to fix NameError: name 'vllm_ops' is not defined (#5009) Letian LiandGitHub 2024-05-23 17:08:58 +01:00
• 6066253296 Marlin 24 prefill performance improvement (about 25% better on average) (#4983) Alexander MatveevandGitHub 2024-05-23 02:39:27 -04:00
• ee3eea0a1b [Misc] Take user preference in attention selector (#4960) Cody YuandGitHub 2024-05-22 15:55:56 -07:00
• a36de682d4 [Minor] Fix small typo in llama.py: QKVParallelLinear -> QuantizationConfig (#4991) Philipp MoritzandGitHub 2024-05-22 15:26:56 -07:00
• eb6d3c264d [Core] Eliminate parallel worker per-step task scheduling overhead (#4894) Nick HillandGitHub 2024-05-22 14:17:27 -07:00
• 97b030005c [Model] LoRA gptbigcode implementation (#3949) raywanbandGitHub 2024-05-23 04:58:59 +08:00
• a3a73ab069 [Misc] Load FP8 kv-cache scaling factors from checkpoints (#4893) Cody YuandGitHub 2024-05-22 13:28:20 -07:00
• 8674f9880e [Kernel] Fixup for CUTLASS kernels in CUDA graphs (#4954) Tyler Michael SmithandGitHub 2024-05-22 10:10:43 -04:00
• c74c913bfb [misc] remove comments that were supposed to be removed (#4977) SangBin ChoandGitHub 2024-05-22 22:02:58 +09:00