Commit Graph

Select branches

Hide Pull Requests

main

Mono Color

• e8ddc08ec8
  [BUG FIX] upgrade fschat version to 0.2.23 (#650) YHPeter 2023-08-02 17:05:59 -0400
• 1b0bd0fe8a
  Add Falcon support (new) (#592) Zhuohan Li 2023-08-02 14:04:39 -0700
• 20044cab7a
  Fix log message in scheduler (#652) Lily Liu 2023-08-02 13:35:10 -0700
• 64f23c2900
  fix baichuan for different position embedding for 7b and 13b models (#643) Song 2023-08-02 13:22:51 +0800
• d4c7755ca8
  fix biachuan-7b tp (#598) Qing 2023-08-02 06:41:36 +0800
• aa39e42c5a
  fix doc (#622) Chaofan Lin 2023-08-01 04:11:57 +0800
• 953f28cf9a
  fix ModuleNotFoundError (#599) Fang li 2023-07-30 11:52:41 +0800
• c0d00f5be6
  [Fix] fix import error of RayWorker (#604) (#605) Xudong Zhang 2023-07-28 14:37:40 +0800
• 58a072be15
  [Fix] Add model sequence length into model config (#575) Zhuohan Li 2023-07-25 23:46:30 -0700
• 82ad323dee
  [Fix] Add chat completion Example and simplify dependencies (#576) Zhuohan Li 2023-07-25 23:45:48 -0700
• df5dd3c68e
  Add Baichuan-7B to README (#494) Zhuohan Li 2023-07-25 15:25:12 -0700
• 2d867b55fa
  fixed tensor parallel is not defined (#564) MoeedDar 2023-07-25 22:16:51 +0100
• d7a1c6d614
  Fix paged attention testing. (#495) Tao Peng 2023-07-25 12:01:56 +0800
• 7d5a155e4a
  [Fix] Fix GPTBigcoder for distributed execution (#503) Zhuohan Li 2023-07-24 18:36:33 -0700
• 1dde34e0f8
  GPTJConfig has no attribute rotary. (#532) leegohi04517 2023-07-25 02:29:30 +0800
• 6fc2a38b11
  Add support for LLaMA-2 (#505) Zhuohan Li 2023-07-20 11:38:27 -0700
• c487a221ee
  Fix bad assert in initialize_cluster if PG already exists (#526) Antoni Baum 2023-07-19 23:17:12 -0700
• 9925c17940
  Ray placement group support (#397) Antoni Baum 2023-07-19 22:49:31 -0700
• 8c4b2592fb
  fix: enable trust-remote-code in api server & benchmark. (#509) Ricardo Lu 2023-07-20 08:06:15 +0800
• cf21a9bd5c
  support trust_remote_code in benchmark (#518) WRH 2023-07-20 08:02:40 +0800
• 16c3e295a8
  fix(ray_utils): ignore re-init error (#465) Massimiliano Pronesti 2023-07-20 02:01:19 +0200
• bda41c70dd
  hotfix attn alibi wo head mapping (#496) Song 2023-07-19 02:31:48 +0800
• 453bafb96f
  Merge pull request #498 from MoeedDar/main Lily Liu 2023-07-18 09:22:56 -0700
• 328d231c17 Fixed old name reference for max_seq_len MoeedDar 2023-07-18 16:47:59 +0100
• b4b195b360
  fix max seq len (#489) Lily Liu 2023-07-17 23:20:20 -0700
• 20b0d88d16
  Add support for baichuan (#365) codethazine 2023-07-17 21:50:55 +0100
• 2bdea7ac11
  [Fix] Fix the condition of max_seq_len (#477) Zhuohan Li 2023-07-17 00:33:48 -0400
• 58df2883cb
  [Doc] Add doc for running vLLM on the cloud (#426) Zhanghao Wu 2023-07-16 13:37:14 -0700
• 6d7d95a70a
  Offload port selection to OS (#467) Zhangir Azerbayev 2023-07-16 02:11:02 -0400
• 96853af5a8
  Optimize MQA Kernel (#452) Zhuohan Li 2023-07-14 20:06:40 -0400
• dbed69058c
  Fix the KeyError when loading bloom-based models (#441) Wen Sun 2023-07-14 12:58:09 +0800
• 7b6ae94059
  add vocab padding for LLama(Support WizardLM) (#411) panda 2023-07-14 11:56:22 +0800
• c6dfc3cdbe
  Fix handling of special tokens in decoding. (#418) xcnick 2023-07-12 23:14:56 +0800
• 51be365143
  fix: freeze pydantic to v1 (#429) Keming 2023-07-12 23:10:55 +0800
• c894836108
  [Model] Add support for GPT-J (#226) Andre Slavescu 2023-07-08 20:55:16 -0400
• 75beba29b5
  Don't try to load training_args.bin (#373) Fazlul Shahriar 2023-07-08 18:26:28 -0400
• ddfdf470ae
  Add trust_remote_code arg to get_config (#405) Woosuk Kwon 2023-07-08 15:24:17 -0700
• b6fbb9a565
  Sort the outputs before return (#402) Woosuk Kwon 2023-07-08 14:48:18 -0700
• 2179e4f4c5
  avoid python list copy in sequence initialization (#401) Lily Liu 2023-07-08 12:42:08 -0700
• a945fcc2ae
  Add trust-remote-code flag to handle remote tokenizers (#364) codethazine 2023-07-07 20:04:58 +0200
• be54f8e5c4
  [Fix] Change /generate response-type to json for non-streaming (#374) Nicolas Frenay 2023-07-06 20:15:17 -0500
• b396cb4998
  fix: only response [DONE] once when streaming response. (#378) Ricardo Lu 2023-07-07 09:08:40 +0800
• 1c395b4eaa
  Bump up the version (#300) Woosuk Kwon 2023-07-04 21:41:53 -0700
• 3d64cf019e
  [Server] use fastchat.model.model_adapter.get_conversation_template method to get model template (#357) akxxsb 2023-07-05 12:39:59 +0800
• 98fe8cb542
  [Server] Add option to specify chat template for chat endpoint (#345) Zhuohan Li 2023-07-03 23:01:56 -0700
• ffa6d2f9f9
  [Docs] Fix typo (#346) Woosuk Kwon 2023-07-03 16:51:47 -0700
• 404422f42e
  [Model] Add support for MPT (#334) Woosuk Kwon 2023-07-03 16:47:53 -0700
• 7717d0838b
  Fix an endless loop issue when engine_step throws a RuntimeError (#339) coolcloudcol 2023-07-04 06:22:28 +0800
• 42e0c1df78
  [Quality] Add CI for formatting (#343) Zhuohan Li 2023-07-03 14:50:56 -0700
• e41f06702c
  Add support for BLOOM (#331) Woosuk Kwon 2023-07-03 13:12:35 -0700
• d6fa1be3a8
  [Quality] Add code formatter and linter (#326) Zhuohan Li 2023-07-03 11:31:55 -0700
• 0ffded812a
  [Fix] Better error message for batched prompts (#342) Zhuohan Li 2023-07-03 09:27:31 -0700
• 0bd2a573a5
  Allow send list of str for the Prompt on openai demo endpoint /v1/completions (#323) Michele Catalano 2023-07-03 18:17:50 +0200
• 49b26e2cec
  feat: add ChatCompletion endpoint in OpenAI demo server. (#330) Ricardo Lu 2023-07-03 13:54:33 +0800
• dafd924c1f
  Raise error for long prompt (#273) Lily Liu 2023-06-30 18:48:49 -0700
• 598dc4b79a
  [Fix] Weight loading for GPTBigCode (#313) Zhuohan Li 2023-06-29 22:14:17 -0700
• 85de093472
  [Fix] Do not pin memory when in WSL (#312) Zhuohan Li 2023-06-29 15:00:21 -0700
• f72297562f
  Add news for the vllm+skypilot example (#314) Zhanghao Wu 2023-06-29 12:32:37 -0700
• 9d27b09d12
  Update README.md (#306) Bayang 2023-06-29 14:52:15 +0100
• 998d9d1509
  [Tokenizer] Add tokenizer mode (#298) Woosuk Kwon 2023-06-28 14:19:22 -0700
• 425040d4c1
  remove floats == 0 comparison (#285) Lily Liu 2023-06-28 14:11:51 -0700
• 4338cc4750
  [Tokenizer] Add an option to specify tokenizer (#284) Woosuk Kwon 2023-06-28 09:46:58 -0700
• bdd6b4c8bc
  Add LLM.set_tokenizer (#283) Jishnu Ray Chowdhury 2023-06-28 02:28:29 -0500
• 2b7d3aca2e
  Update setup.py (#282) Cody Yu 2023-06-27 14:34:23 -0700
• 4026a049d3
  expand coverage of gpt2 model loading (#271) twaka 2023-06-27 22:27:41 +0900
• 43710e8d09
  [Fix] Fix default port number in benchmark scripts (#265) Zhuohan Li 2023-06-26 13:15:35 -0700
• 526df28fb2
  [BugFix] Fix a bug in counting running sequences (#266) Woosuk Kwon 2023-06-26 13:09:02 -0700
• 2cf1a333b6
  [Doc] Documentation for distributed inference (#261) Zhuohan Li 2023-06-26 11:34:23 -0700
• 0b7db411b5
  [Bug] Fix the OOM condition for CPU cache (#260) Zhuohan Li 2023-06-26 11:16:13 -0700
• 471a7a4566
  Compatible with Decapoda Research llama hf version (#251) BasicCoder 2023-06-27 00:23:57 +0800
• 6214dd6ce9
  Update README.md (#236) Lianmin Zheng 2023-06-25 16:58:06 -0700
• 0603379863
  fix wrong using getattr to get dict value (#232) metacryptom 2023-06-25 13:00:24 +0800
• 665c48963b
  [Docs] Add GPTBigCode to supported models (#213) Woosuk Kwon 2023-06-22 15:05:11 -0700
• 298695b766
  GPTBigCode (StarCoder, SantaCoder Support) (#209) Michael Feil 2023-06-22 19:49:27 +0200
• 83658c8ace
  Bump up version to 0.1.1 (#204) Zhuohan Li 2023-06-22 15:33:32 +0800
• 1d24ccb96c
  [Fix] Better error message when there is OOM during cache initialization (#203) Zhuohan Li 2023-06-22 15:30:06 +0800
• 14f0b39cda
  [Bugfix] Fix a bug in RequestOutput.finished (#202) Woosuk Kwon 2023-06-22 00:17:24 -0700
• 2e0d314384
  fix-ray (#193) Zhuohan Li 2023-06-22 00:21:41 +0800
• 67d96c29fb
  Use slow tokenizer for open llama models (#168) Woosuk Kwon 2023-06-19 23:19:47 -0700
• 033f5c78f5
  Remove e.g. in README (#167) Zhuohan Li 2023-06-20 14:00:28 +0800
• 794e578de0
  [Minor] Fix URLs (#166) Woosuk Kwon 2023-06-19 22:57:14 -0700
• caddfc14c1
  [Minor] Fix icons in doc (#165) Woosuk Kwon 2023-06-19 20:35:38 -0700
• fc72e39de3
  Change image urls (#164) Zhuohan Li 2023-06-20 11:15:15 +0800
• b7e62d3454
  Fix repo & documentation URLs (#163) Woosuk Kwon 2023-06-19 20:03:40 -0700
• 364536acd1
  [Docs] Minor fix (#162) Woosuk Kwon 2023-06-19 19:58:23 -0700
• 0b32a987dd
  Add and list supported models in README (#161) Zhuohan Li 2023-06-20 10:57:46 +0800
• 570fb2e9cc
  [PyPI] Fix package info in setup.py (#158) Woosuk Kwon 2023-06-19 18:05:01 -0700
• a255885f83
  Add logo and polish readme (#156) Zhuohan Li 2023-06-19 16:31:13 +0800
• 5822ede66e
  Add performance figures for dark mode (#160) Woosuk Kwon 2023-06-18 23:46:24 -0700
• 0370afa2e5
  Remove benchmark_async_llm_server.py (#155) Zhuohan Li 2023-06-19 11:12:37 +0800
• 7e2a913c64
  [Minor] Fix CompletionOutput.__repr__ (#157) Woosuk Kwon 2023-06-18 19:58:25 -0700
• 3f92038b99
  Add comments on swap space (#154) Woosuk Kwon 2023-06-18 11:39:35 -0700
• dcda03b4cb
  Write README and front page of doc (#147) Woosuk Kwon 2023-06-18 03:19:38 -0700
• bf5f121c02
  Reduce GPU memory utilization to make sure OOM doesn't happen (#153) Zhuohan Li 2023-06-18 17:33:50 +0800
• bec7b2dc26
  Add quickstart guide (#148) Zhuohan Li 2023-06-18 01:26:12 +0800
• 0b98ba15c7
  Change the name to vLLM (#150) Woosuk Kwon 2023-06-17 03:07:40 -0700
• e5464ee484
  Rename servers to engines (#152) Zhuohan Li 2023-06-17 17:25:21 +0800
• bab8f3dd0d
  [Minor] Fix benchmark_throughput.py (#151) Woosuk Kwon 2023-06-16 21:00:52 -0700
• eedb46bf03
  Rename servers and change port numbers to reduce confusion (#149) Zhuohan Li 2023-06-17 00:13:02 +0800
• 311490a720
  Add script for benchmarking serving throughput (#145) Woosuk Kwon 2023-06-14 19:55:38 -0700