Commit graph

Select branches

Hide pull requests

main

Mono Color

• b2f62a8561
  Add batch 4 perf test (#11355) Wenjing Margaret Mao 2024-06-20 09:48:52 +0800
• a721c1ae43
  minor fix of ragflow_quickstart.md (#11364) Shengsheng Huang 2024-06-19 22:30:33 +0800
• 13727635e8
  revise ragflow quickstart (#11363) Shengsheng Huang 2024-06-19 22:24:31 +0800
• 5283df0078
  LLM: Add RAGFlow with Ollama Example QuickStart (#11338) Zijie Li 2024-06-19 20:00:50 +0800
• ae452688c2
  Add NPU HF example (#11358) Zijie Li 2024-06-19 18:07:28 +0800
• 1eb884a249
  IPEX Duplicate importer V2 (#11310) Qiyuan Gong 2024-06-19 16:29:19 +0800
• 271d82a4fc
  Update readme (#11357) Jason Dai 2024-06-19 10:05:42 +0800
• ae7b662ed2
  add fp16 NPU Linear support and fix intel_npu_acceleration_library version 1.0 support (#11352) Yishuo Wang 2024-06-19 09:14:59 +0800
• c44b1942ed
  fix mistral for transformers>=4.39 (#11191) Guoqiong Song 2024-06-18 13:39:35 -0700
• 67a1e05876
  Remove zero3 context manager from LoRA (#11346) Heyang Sun 2024-06-18 17:24:43 +0800
• f6cd628cd8
  Fix script usage in vLLM CPU Quickstart (#11353) Xiangyu Tian 2024-06-18 16:50:48 +0800
• ef9f740801
  Docs: Fix CPU Serving Docker README (#11351) Xiangyu Tian 2024-06-18 16:27:51 +0800
• c9b4cadd81
  fix vLLM/docker issues (#11348) Guancheng Fu 2024-06-18 16:23:53 +0800
• 83082e5cc7
  add initial support for intel npu acceleration library (#11347) Yishuo Wang 2024-06-18 16:07:16 +0800
• 694912698e
  Upgrade scikit-learn to 1.5.0 to fix dependabot issue (#11349) Shaojun Liu 2024-06-18 15:47:25 +0800
• 44f22cba70
  add config and default value (#11344) hxsz1997 2024-06-18 15:28:57 +0800
• 1f39bb84c7
  update readthedocs perf data (#11345) Shengsheng Huang 2024-06-18 13:23:47 +0800
• 00f322d8ee
  Finetune ChatGLM with Deepspeed Zero3 LoRA (#11314) Heyang Sun 2024-06-18 12:31:26 +0800
• 5dad33e5af
  Support fp8_e4m3 scale search (#11339) Yina Chen 2024-06-18 11:47:43 +0800
• e50c890e1f
  Support finishing PP inference once eos_token_id is found (#11336) binbin Deng 2024-06-18 09:55:40 +0800
• de4bb97b4f
  Remove accelerate 0.23.0 install command in readme and docker (#11333) Qiyuan Gong 2024-06-17 17:52:12 +0800
• ef4b6519fb
  Add phi-3 model support for pipeline parallel inference (#11334) SONG Ge 2024-06-17 17:44:24 +0800
• 99b309928b
  Add lookahead in test_api: transformer_int4_fp16_gpu (#11337) hxsz1997 2024-06-17 17:41:41 +0800
• bc4bafffc7
  Update README.md (#11335) Jason Dai 2024-06-17 16:24:23 +0800
• 5d7c9bf901
  Upgrade accelerate to 0.23.0 (#11331) Qiyuan Gong 2024-06-17 15:03:11 +0800
• 183e0c6cf5
  glm-4v-9b support (#11327) Xin Qiu 2024-06-17 13:52:37 +0800
• bca5cbd96c
  Modify arc nightly perf to fp16 (#11275) Wenjing Margaret Mao 2024-06-17 13:47:22 +0800
• a2a5890b48
  Make manually-triggered perf test able to choose which test to run (#11324) Yuwen Hu 2024-06-17 10:23:13 +0800
• 1978f63f6b
  Fix igpu performance guide regarding html generation (#11328) Yuwen Hu 2024-06-17 10:21:30 +0800
• 6ea1e71af0
  Update PP inference benchmark script (#11323) binbin Deng 2024-06-17 09:59:36 +0800
• be00380f1a
  Fix pipeline parallel inference past_key_value error in Baichuan (#11318) SONG Ge 2024-06-17 09:29:32 +0800
• 0af0102e61
  Add quantization scale search switch (#11326) Yina Chen 2024-06-14 18:46:52 +0800
• 8a3247ac71
  support batch forward for q4_k, q6_k (#11325) Ruonan Wang 2024-06-14 18:25:50 +0800
• e8dd8e97ef
  fix chatglm lookahead on ARC (#11320) Yishuo Wang 2024-06-14 16:26:11 +0800
• f5ef94046e
  exclude dolly-v2-12b for arc perf test (#11315) Shaojun Liu 2024-06-14 15:35:56 +0800
• 77809be946
  Install packages for ipex-llm-serving-cpu docker image (#11321) Shaojun Liu 2024-06-14 15:26:01 +0800
• 4359ab3172
  LLM: Add /generate_stream endpoint for Pipeline-Parallel-FastAPI example (#11187) Xiangyu Tian 2024-06-14 15:15:32 +0800
• 9e4d87a696
  Langchain-chatchat QuickStart small link fix (#11317) Yuwen Hu 2024-06-14 14:02:17 +0800
• 0e7a31a09c
  ChatGLM Examples Restructure regarding Installation Steps (#11285) Jin Qiao 2024-06-14 12:37:05 +0800
• 91965b5d05
  add glm_sdpa back to fix chatglm-6b (#11313) Yishuo Wang 2024-06-14 10:31:43 +0800
• 7f65836cb9
  fix chatglm2/3-32k/128k fp16 (#11311) Yishuo Wang 2024-06-14 09:58:07 +0800
• 1b0c4c8cb8
  use new rotary two in chatglm4 (#11312) Xin Qiu 2024-06-13 19:02:18 +0800
• f1410d6823
  refactor chatglm4 (#11301) Xin Qiu 2024-06-13 18:06:04 +0800
• 5e25766855
  fix and optimize chatglm2-32k and chatglm3-128k (#11306) Yishuo Wang 2024-06-13 17:37:58 +0800
• 60cb1dac7c
  Support PP for qwen1.5 (#11300) binbin Deng 2024-06-13 17:35:24 +0800
• f97cce2642
  Fix import error of ds autotp (#11307) binbin Deng 2024-06-13 16:22:52 +0800
• 3682c6a979
  add glm4 and qwen2 to igpu perf (#11304) Jin Qiao 2024-06-13 16:16:35 +0800
• a24666b8f3
  fix chatglm3-6b-32k (#11303) Yishuo Wang 2024-06-13 16:01:34 +0800
• 9760ffc256
  Fix SDLe CT222 Vulnerabilities (#11237) Shaojun Liu 2024-06-13 15:31:22 +0800
• bfab294f08
  Update langchain-chatchat QuickStart to include Core Ultra iGPU Linux Guide (#11302) Yuwen Hu 2024-06-13 15:09:55 +0800
• 84f04087fb
  Add intelanalytics/ipex-llm:sources image for OSPDT (#11296) Shaojun Liu 2024-06-13 14:29:14 +0800
• 01fe0fc1a2
  refactor chatglm2/3 (#11290) Yishuo Wang 2024-06-13 12:22:58 +0800
• ea372cc472
  update demos section (#11298) Shengsheng Huang 2024-06-13 11:58:19 +0800
• 57a023aadc
  Fix vllm tp (#11297) Guancheng Fu 2024-06-13 10:47:48 +0800
• 986af21896
  fix perf test(#11295) Ruonan Wang 2024-06-13 10:35:48 +0800
• 220151e2a1
  Refactor pipeline parallel multi-stage implementation (#11286) binbin Deng 2024-06-13 10:00:23 +0800
• 14b1e6b699
  Fix gguf_q4k (#11293) Ruonan Wang 2024-06-12 20:43:08 +0800
• 8edcdeb0e7
  Fix bug that torch.ops.torch_ipex.matmul_bias_out cannot work on Linux MTL for short input (#11292) Yuwen Hu 2024-06-12 19:12:57 +0800
• b61f6e3ab1
  Add update_parent_folder for nightly_perf_test (#11287) Wenjing Margaret Mao 2024-06-12 17:58:13 +0800
• 2e75bbccf9
  Add more control arguments for benchmark_vllm_throughput (#11291) Guancheng Fu 2024-06-12 17:43:06 +0800
• 592f7aa61e
  Refine glm1-4 sdp (#11276) Xin Qiu 2024-06-12 17:11:56 +0800
• cffb932f05
  Expose timeout for streamer for fastchat worker (#11288) Yuwen Hu 2024-06-12 17:02:40 +0800
• d99423b75a
  Readme demo (#11283) Shengsheng Huang 2024-06-12 17:01:53 +0800
• e7a4e2296f
  Add Stable Diffusion examples on GPU and CPU (#11166) ivy-lv11 2024-06-12 16:33:25 +0800
• f224e98297
  Add GLM-4 CPU example (#11223) Jin Qiao 2024-06-12 15:30:51 +0800
• 40fc8704c4
  Add GPU example for GLM-4 (#11267) Zijie Li 2024-06-12 14:29:50 +0800
• 0d9cc9c106
  Remove duplicate check for ipex (#11281) Qiyuan Gong 2024-06-12 13:52:02 +0800
• 10e480ee96
  refactor internlm and internlm2 (#11274) Yishuo Wang 2024-06-11 14:19:19 +0800
• fac49f15e3
  Remove manual importing ipex in all-in-one benchmark (#11272) Yuwen Hu 2024-06-11 09:32:13 +0800
• 70b17c87be
  Merge multiple batches (#11264) Wenjing Margaret Mao 2024-06-07 18:38:45 +0800
• 4b07712fd8
  LLM: Fix vLLM CPU model convert mismatch (#11254) Xiangyu Tian 2024-06-07 15:54:34 +0800
• 42fab480ea
  support stablm2 12b (#11265) Yishuo Wang 2024-06-07 15:46:00 +0800
• dbc3c2d72d
  glm4 sdp (#11253) Xin Qiu 2024-06-07 15:42:23 +0800
• 151fcf37bb
  check devie name in use_flash_attention (#11263) Xin Qiu 2024-06-07 15:07:47 +0800
• 2623944604
  qwen2 sdpa small fix (#11261) Yishuo Wang 2024-06-07 14:42:18 +0800
• ea0d03fd28
  Refactor baichuan1 7B and 13B (#11258) Yishuo Wang 2024-06-07 14:29:20 +0800
• 1aa9c9597a
  Avoid duplicate import in IPEX auto importer (#11227) Qiyuan Gong 2024-06-07 14:08:00 +0800
• 6f2684e5c9
  Update pp llama.py to save memory (#11233) Wang, Jian4 2024-06-07 13:18:16 +0800
• ef8e9b2ecd
  Refactor qwen2 moe (#11244) Yishuo Wang 2024-06-07 13:14:54 +0800
• 7b753dc8ca
  Update sample output for HF Qwen2 GPU and CPU (#11257) Zijie Li 2024-06-07 11:36:22 +0800
• b7948671de
  [WIP] Add look up table in 1st token stage (#11193) Zhao Changmin 2024-06-07 10:51:05 +0800
• 375174af33
  Small qwen2 link fix (#11255) Yuwen Hu 2024-06-07 10:37:32 +0800
• 8c36b5bdde
  Add qwen2 example (#11252) Yuwen Hu 2024-06-07 10:29:33 +0800
• 85df5e7699
  fix nightly perf test (#11251) Shaojun Liu 2024-06-07 09:33:14 +0800
• 2f809116e2
  optimize Chatglm4 (#11239) Xin Qiu 2024-06-06 18:25:20 +0800
• eeffeeb2e2
  fix benchmark script(#11243) Guancheng Fu 2024-06-06 17:44:19 +0800
• 8aabb5bac7
  Enable CodeQL Check for CT39 (#11242) Shaojun Liu 2024-06-06 17:41:12 +0800
• b6234eb4e2
  Add task in allinone (#11226) hxsz1997 2024-06-06 17:22:40 +0800
• c825a7e1e9
  change the workflow file to test ftp (#11241) Wenjing Margaret Mao 2024-06-06 16:53:19 +0800
• 2e4ccd541c
  fix qwen2 cpu (#11240) Yishuo Wang 2024-06-06 16:24:19 +0800
• e738ec38f4
  disable quantize kv in specific qwen model (#11238) Yishuo Wang 2024-06-06 14:08:39 +0800
• c4e5806e01
  add latest optimization in starcoder2 (#11236) Yishuo Wang 2024-06-06 14:02:17 +0800
• ba27e750b1
  refactor yuan2 (#11235) Yishuo Wang 2024-06-06 13:17:54 +0800
• 6be24fdd28
  OSPDT: add tpp licenses (#11165) Shaojun Liu 2024-06-06 10:59:06 +0800
• 09c6780d0c
  phi-2 transformers 4.37 (#11161) Guoqiong Song 2024-06-05 13:36:41 -0700
• f6d5c6af78
  fix issue 1407 (#11171) Guoqiong Song 2024-06-05 13:35:57 -0700
• bfa1367149
  Add CPU and GPU example for MiniCPM (#11202) Zijie Li 2024-06-05 18:09:53 +0800
• a27a559650
  Add some information in FAQ to help users solve "RuntimeError: could not create a primitive" error on Windows (#11221) Xu, Shuo 2024-06-05 17:57:42 +0800
• af96579c76
  Update installation guide for pipeline parallel inference (#11224) Yuwen Hu 2024-06-05 17:54:29 +0800
• ed67435491
  Support Fp6 k in ipex-llm (#11222) Yina Chen 2024-06-05 17:34:36 +0800