Code

New functions added#13

Code

Open

SANKHA1 wants to merge 220 commits intoSANKHA1:sankha_branchSANKHA1/ipex-llm:sankha_branchfrom intel-analytics:mainintel-analytics/ipex-llm:mainCopy head branch name to clipboard

+11,826-10,138

Commits

Commits on Nov 4, 2024

• Fix performance tests regarding trl version (#12319)

  Show description for 94ce447
  Oscilloscope98

  authored
  94ce447

• Qwen layernorm as input (#12309)

  Show description for c8679ad
  hkvision

  authored
  c8679ad

• [NPU pipeline] update cmake usage of pipeline (#12320)

  rnwang04

  authored
  8fe01c9

• Perf test further fix regarding trl version (#12321)

  Oscilloscope98

  authored
  4644cb6

• Doc: update harness readme (#12324)

  cranechu0131

  authored
  a01371f

• [NPU L0] Add layernorm weight as const / input setting (#12322)

  plusbang

  authored
  5ee6f97

• Add transformers_int4_npu_pipeline_win in all-in-one benchmark (#12325)

  Show description for e54af44
  ch1y0q

  authored
  e54af44

• [NPU] Add env to disable compile opt (#12330)

  Show description for 94c4ce3
  cyita

  authored
  94c4ce3

• Add chatglm2&3 fuse mlp (#12328)

  Show description for 1b637e4
  leonardozcm

  authored
  1b637e4

• Add initial support for LNL nightly performance tests (#12326)

  Show description for 522cdf8
  Oscilloscope98

  authored
  522cdf8

• Small fix to LNL performance tests (#12331)

  Oscilloscope98

  authored
  e2adc97

Commits on Nov 5, 2024

• update benchmark readme (#12323)

  Show description for 45b0d37
  lzivan

  authored
  45b0d37

• Small fix to LNL performance tests (#12333)

  Oscilloscope98

  authored
  923d696

• Limit trl version in example (#12332)

  Show description for 82a61b5
  JinBridger

  authored
  82a61b5

• [NPU] Llama3, Qwen2 1.5b, MiniCPM 1/2B groupwise support (#12327)

  Show description for d872639
  cyita

  authored
  d872639

• fix chatglm2 cpu ut (#12336)

  leonardozcm

  authored
  8e9a3a1

• Add dummy model in iGPU perf (#12341)

  Show description for 7240c28
  JinBridger

  authored
  7240c28

Commits on Nov 6, 2024

• Replace gradio_web_server.patch to adjust webui (#12329)

  Show description for 899a303
  

  ATMxsp01

  and

  ATMxsp01

  authored
  899a303

• [NPU] Hot fix of load_low_bit (#12344)

  plusbang

  authored
  69e3a56

• Add basic glm4v support (#12345)

  MeouSker77

  authored
  c8b7265

• optimize glm4v's vision part (#12346)

  MeouSker77

  authored
  e23ef7d

• Add MiniCPM-V-2_6 to arc perf test (#12349)

  JinBridger

  authored
  d984c06

• llama 3.1/3.2 support compresskv (#12347)

  Show description for f24352a
  cyita

  authored
  f24352a

• fix three NPU benchmark issues (#12350)

  Show description for c267355
  rnwang04

  authored
  c267355

• Small optimization to glm4 models (#12351)

  Oscilloscope98

  authored
  872a744

• [NPU] Add Optimized Support for Llama3.2-1B/3B on NPU (#12339)

  Show description for a7b6668
  sgwhat

  authored
  a7b6668

Commits on Nov 7, 2024

• add minicpm-v models to transformers_int4_npu_win api (#12352)

  Show description for 79f2877
  JinheTang

  authored
  79f2877

• [NPU] acclib llama3.2 support groupwise (#12355)

  Show description for d880e53
  cyita

  authored
  d880e53

• Update Readme for FastChat docker demo (#12354)

  Show description for ce0c6ae
  

  ATMxsp01

  and

  ATMxsp01

  authored
  ce0c6ae

• Add troubleshootings for ollama and llama.cpp (#12358)

  Show description for 71ea539
  JinheTang

  authored
  71ea539

• small improvement (#12359)

  MeouSker77

  authored
  ad68c56

• Update install_linux_gpu.md (#12353)

  qiuxin2012

  authored
  520af4e

• Add fused mlp optimizations to glm4 models (#12360)

  Show description for 1a6cbc4
  Oscilloscope98

  authored
  1a6cbc4

• Small fix to all-in-one benchmark (#12362)

  Oscilloscope98

  authored
  8fe294e

Commits on Nov 8, 2024

• update linux installation doc (#12365)

  Show description for 7ef7696
  qiuxin2012

  authored
  7ef7696

• [NPU L0] Support llama3.2 in L0 pipeline (#12361)

  plusbang

  authored
  812d5cc

• [NPU] Support Baichuan groupwise & gw code refactor (#12337)

  Show description for b2e69a8
  cyita

  authored
  b2e69a8

• fix ipex 2.3 bug (#12366)

  MeouSker77

  authored
  51f7f87

• Update fastchat demo script (#12367)

  Show description for fad15c8
  liu-shaojun

  authored
  fad15c8

• Fix trl version and padding in trl qlora example (#12368)

  Show description for 2dfcc36
  qiyuangong

  authored
  2dfcc36

• optimize glm4v vision attention (#12369)

  MeouSker77

  authored
  dc34e8c

Commits on Nov 11, 2024

• Add fused_mlp to glm4v models (#12378)

  Oscilloscope98

  authored
  e091893

• Update oneccl-binding.patch (#12377)

  Show description for c92d76b
  liu-shaojun

  authored
  c92d76b

Commits on Nov 12, 2024

• Update llama-cpp docker usage (#12387)

  hzjane

  authored
  85c9279

• Support vpm and resampler module of minicpm-v on NPU (#12375)

  plusbang

  authored
  7a97fbb

• [NPU] Update qwen2 compile config (#12383)

  Show description for 6bf5a8c
  rnwang04

  authored
  6bf5a8c

• Decouple the openwebui and the ollama. in inference-cpp-xpu dockerfile (#12382)

  Show description for 4376fde
  ACupofAir

  authored
  4376fde

• Upgrade to vllm 0.6.2 (#12338)

  Show description for 0ee54fc
  

  gc-fu

  and

  xiangyuT

  authored
  0ee54fc

• changed inference-cpp/Dockerfile (#12386)

  Show description for dd8964b
  

  3 peopleauthored
  dd8964b

• minor fix (#12389)

  liu-shaojun

  authored
  2715247

Commits on Nov 13, 2024

• Fix llava with multi-image inputs (#12384)

  Oscilloscope98

  authored
  1158f91

• qwen prefill attn_mask type fp16 (#12394)

  cyita

  authored
  9220bab

• [NPU] Qwen prefill attn_mask type hotfix (#12395)

  Show description for d6d63d6
  cyita

  authored
  d6d63d6

• use new q4_0 batch kernel (#12396)

  MeouSker77

  authored
  00fce5c

Commits on Nov 14, 2024

• small fix (#12397)

  cyita

  authored
  59b01fa

• Update readme & doc for the vllm upgrade to v0.6.2 (#12399)

  Show description for 6726b19
  

  ATMxsp01

  and

  ATMxsp01

  authored
  6726b19

• Add initial support for modeling_xlm encoder on NPU (#12393)

  Show description for d2cbcb0
  sgwhat

  authored
  d2cbcb0

• Add padding_token=eos_token for GPU trl QLora example (#12398)

  Show description for 7e50ff1
  qiyuangong

  authored
  7e50ff1

• [NPU] change attention_mask to fp16 (#12400)

  plusbang

  authored
  d4d9494

Commits on Nov 15, 2024

• fix npu pipeline workflow (#12404)

  rnwang04

  authored
  548dec5

• Tiny doc fix (#12405)

  Oscilloscope98

  authored
  d1cde7f

• fix workflow again (#12406)

  Show description for fcc0fa7
  rnwang04

  authored
  fcc0fa7

• fix again (#12407)

  rnwang04

  authored
  6c5e8fc

• update batch kernel condition (#12408)

  MeouSker77

  authored
  3d5fbf2

Commits on Nov 18, 2024

• Add missing arguments in pipeline parallel generate method (#12142)

  Show description for d2c821d
  notsyncing

  authored
  d2c821d

• Support performance mode of GLM4 model (#12401)

  Show description for a69395f
  

  Oscilloscope98

  and

  cyita

  authored
  a69395f

Commits on Nov 19, 2024

• Update benchmark_vllm_throughput.py (#12414)

  gc-fu

  authored
  d6057f6

• Add install_windows_gpu.zh-CN.md and install_linux_gpu.zh-CN.md (#12409)

  Show description for a9cb70a
  joan726

  authored
  a9cb70a

• Fix speech_paraformer issue with unexpected changes (#12416)

  Show description for ff3f7cb
  sgwhat

  authored
  ff3f7cb

Commits on Nov 20, 2024

• Add multimodal benchmark (#12415)

  Show description for 1bfcbc0
  hzjane

  authored
  1bfcbc0

• [NPU] dump prefill IR for further C++ solution (#12402)

  Show description for 54c62fe
  rnwang04

  authored
  54c62fe

• add Stable diffusion examples (#12418)

  Show description for d2a37b6
  JinheTang

  authored
  d2a37b6

Commits on Nov 21, 2024

• Initial NPU C++ Example (#12417)

  Show description for 7288c75
  rnwang04

  authored
  7288c75

• update batch kernel condition (#12421)

  MeouSker77

  authored
  145e8b4

• add optimization to openjourney (#12423)

  Show description for 7e0a840
  JinheTang

  authored
  7e0a840

• Optimize with new batch kernel when batch_size=1 on LNL (#12419)

  Show description for 8fdc36c
  Oscilloscope98

  authored
  8fdc36c

• small fix of cpp readme(#12425)

  rnwang04

  authored
  2935e97

• Upgrade denpendency for xpu_lnl and xpu_arl option (#12424)

  Oscilloscope98

  authored
  e61ae88

Commits on Nov 22, 2024

• Update english prompt to 34k (#12429)

  liu-shaojun

  authored
  c089b6c

• New convert support for C++ NPU (#12430)

  Show description for 4ffa6c7
  rnwang04

  authored
  4ffa6c7

• support Llama2-7B / Llama3-8B for NPU C++ (#12431)

  Show description for 0819fad
  rnwang04

  authored
  0819fad

Commits on Nov 25, 2024

• Support minicpm for NPU C++ (#12434)

  Show description for f414053
  rnwang04

  authored
  f414053

• fix and optimize sd (#12436)

  MeouSker77

  authored
  be132c4

• small change (#12439)

  MeouSker77

  authored
  8164aed

• add chinese prompt troubleshooting for npu cpp examples (#12437)

  Show description for b633fbf
  JinheTang

  authored
  b633fbf

• Support qwen2.5 3B for NPU & update related examples (#12438)

  Show description for b9abb8a
  rnwang04

  authored
  b9abb8a

• optimize sdxl again (#12441)

  MeouSker77

  authored
  cdd41f5

Commits on Nov 26, 2024

• Add support of llama3.2 for NPU C++ (#12442)

  Show description for 0e23bd7
  rnwang04

  authored
  0e23bd7

• add sdxl and lora-lcm optimization (#12444)

  Show description for 66bd7ab
  JinheTang

  authored
  66bd7ab

• Optimize first token of C++ NPU by adding npu_dpu_groups (#12443)

  Show description for 52c17fe
  rnwang04

  authored
  52c17fe

• update serving image runtime (#12433)

  pepijndevos

  authored
  71e1f11

• Fix abnormal output for Qwen2-7B when sym_int8 (#12446)

  Oscilloscope98

  authored
  303b104

• [NPU] further fix of qwen2 int8 pipeline & C++ (#12449)

  Show description for 24b46b2
  rnwang04

  authored
  24b46b2

• Update LangChain examples to use upstream (#12388)

  Show description for c2efa26
  JinBridger

  authored
  c2efa26

• [NPU] Support GW for NPU C++ (#12450)

  rnwang04

  authored
  7b40f9b

Commits on Nov 27, 2024

• update vllm-docker-quick-start for vllm0.6.2 (#12392)

  Show description for cb7b089
  ACupofAir

  authored
  cb7b089

• Fix (#12390)

  gc-fu

  authored
  8331875

• [NPU] optimize qwen2 prefill performance for C++ (#12451)

  rnwang04

  authored
  f8c2bb2

• Remove env variable BIGDL_LLM_XMX_DISABLED in documentation (#12445)

  Show description for acd77d9
  cranechu0131

  authored
  acd77d9

• Small update to LangChain examples readme (#12452)

  Oscilloscope98

  authored
  effb9bb

• update transformers version in example of glm4 (#12453)

  Show description for ce6fcaa
  cranechu0131

  authored
  ce6fcaa

• [NPU] Add L0 support for NPU C++ (#12454)

  Show description for 281c9b0
  rnwang04

  authored
  281c9b0

• fix glm4-9b overflow (#12455)

  MeouSker77

  authored
  6f3441b

• Small fix in llama.cpp troubleshooting guide (#12457)

  Oscilloscope98

  authored
  a2272b7

• [NPU] Update C++ L0 (#12458)

  Show description for b29da30
  rnwang04

  authored
  b29da30

Commits on Nov 28, 2024

• remove nf4 unsupport comment in cpu finetuning (#12460)

  Show description for d272f6b
  

  Uxito-Ada

  and

  Ariadne330

  authored
  d272f6b

• [NPU] Add env to enable scale search (#12462)

  Show description for 1b533a1
  cyita

  authored
  1b533a1

• [NPU] update fused layers for GW (#12459)

  Show description for 490bb0c
  rnwang04

  authored
  490bb0c

Commits on Nov 29, 2024

• Integrate NPU C++ imple into ipex-llm (#12461)

  plusbang

  authored
  14d8d3d

• [NPU C++] Update model support & examples & benchmark (#12466)

  plusbang

  authored
  c911026

• Hotfix of benchmark script (#12467)

  plusbang

  authored
  f99f188

Commits on Dec 2, 2024

• Support imatrix-guided quantization for NPU CW (#12468)

  Show description for 4b6c316
  rnwang04

  authored
  4b6c316

• add vLLM glm4 fix (#12474)

  gc-fu

  authored
  59bd4a2

• [NPU]Fix eos_token setting (#12475)

  plusbang

  authored
  54d9a59

• Fix MiniCPM-V models running on NPU (#12478)

  JinBridger

  authored
  31c69a8

• Add NPU QuickStart & update example links (#12470)

  Show description for aee9acb
  Oscilloscope98

  authored
  aee9acb

• Add release support for option xpu_arc (#12422)

  Show description for b2e56a2
  Oscilloscope98

  authored
  b2e56a2

• [NPU] Remove hard code (#12479)

  Oscilloscope98

  authored
  26adb82

Commits on Dec 3, 2024

• [NPU] update save-load API usage (#12473)

  plusbang

  authored
  ab01753

• small fix of imatrix (#12480)

  rnwang04

  authored
  598603b

• Fix MiniCPM-V-2_6 running on NPU (#12486)

  JinBridger

  authored
  5fe7667

• Fix NPU LLM example save/load tokenizer (#12485)

  JinBridger

  authored
  7082844

• [NPU] Support streaming in Python (cpp backend) (#12488)

  Show description for 4ac66db
  Oscilloscope98

  authored
  4ac66db

• Update README.md (#12489)

  jason-dai

  authored
  80f15e4

• Hotfix of BCE-Emdedding model (#12490)

  plusbang

  authored
  c592844

Commits on Dec 4, 2024

• optimize qwen2_vl multiple image input or video input (#12487)

  MeouSker77

  authored
  5629fdd

• [NPU] Support split lm_head for Qwen2 with CPP (#12491)

  Show description for ef4028a
  Oscilloscope98

  authored
  ef4028a

• Support hf generate (#12477)

  Show description for 7ff4533
  hkvision

  authored
  7ff4533

• small fix (#12493)

  MeouSker77

  authored
  e0bf005

• Added cross-links (#12494)

  Show description for ae9c215
  joan726

  authored
  ae9c215

• optimize minicpm (#12496)

  MeouSker77

  authored
  a9e3f7f

• Update streaming in npu examples (#12495)

  Show description for ffa9a9e
  

  cranechu0131

  and

  Oscilloscope98

  authored
  ffa9a9e

• Fix hf generate for llama3.2 (#12497)

  Show description for 7d27f13
  hkvision

  authored
  7d27f13

• Support save/load model for hf generate (#12499)

  Show description for b89ea1b
  hkvision

  authored
  b89ea1b

• Update save/load comments (#12500)

  hkvision

  authored
  d8b14a6

• Small fix for NPU Python cpp simple generate regarding eos tokens (#12501)

  Oscilloscope98

  authored
  84f1c4a

Commits on Dec 5, 2024

• [NPU] Fix load-low-bit benchmark script (#12502)

  plusbang

  authored
  f56a111

• Add NPU demo gif to main readme (#12503)

  Show description for 727f299
  Oscilloscope98

  authored
  727f299

• fix readme for npu cpp examples and llama.cpp (#12505)

  Show description for 5e1416c
  JinheTang

  authored
  5e1416c

• Update README.md (#12507)

  jason-dai

  authored
  0a3eda0

• Small fixes to main readme (#12508)

  Oscilloscope98

  authored
  60bafab

• [NPU] initial support of asym_int4_rtn (#12484)

  Show description for 49ab897
  rnwang04

  authored
  49ab897

• [NPU] Fix hf generate with save/load generation config for Python (cpp backend) (#12509)

  Show description for 0918d3b
  Oscilloscope98

  authored
  0918d3b

Commits on Dec 6, 2024

• [NPU C++] Update example with conversation mode support (#12510)

  plusbang

  authored
  12c7897

Commits on Dec 9, 2024

• [NPU] Support glm-edge models (#12511)

  plusbang

  authored
  ea55235

Commits on Dec 10, 2024

• vllm oneccl upgrade to b9 (#12520)

  hzjane

  authored
  922958c

Commits on Dec 11, 2024

• support new model (#12523)

  MeouSker77

  authored
  77404d2

• [NPU] Support repetition penalty for simple generate, Python (cpp backend) (#12522)

  Show description for 68f2873
  Oscilloscope98

  authored
  68f2873

• support hqq (#12518)

  Show description for 588bfa2
  rnwang04

  authored
  588bfa2

• [NPU] fix transpose_value = False for NPU optimize_model=True (#12525)

  rnwang04

  authored
  41ef497

• All-in-one Benchmark run.py: Ignore error if import BenchmarkWrapper failed. (#12526)

  ATMxsp01

  authored
  fd9cf76

• [NPU] Fix minicpm-2B error (#12527)

  plusbang

  authored
  509bdb4

Commits on Dec 12, 2024

• [NPU] Update glm-edge support in docs (#12529)

  plusbang

  authored
  6fc27da

• Enable use_batch_forward Optimization on Battlemage GPU (#12516)

  Show description for 2cce896
  liu-shaojun

  authored
  2cce896

• [NPU] Update C++ example with repetition_penalty & update Python code accordingly (#12528)

  Show description for dbaf4ab
  Oscilloscope98

  authored
  dbaf4ab

• add basic glm-edge support (#12531)

  MeouSker77

  authored
  3e0823d

• add basic glm-edge-v support (#12533)

  MeouSker77

  authored
  ffce86d

• [NPU] Fix abnormal output with latest driver (#12530)

  plusbang

  authored
  f36c236

Commits on Dec 13, 2024

• Small fix to GPU installation guide (#12536)

  Oscilloscope98

  authored
  b747f3f

• torch 2.3 inference docker (#12517)

  Show description for fa261b8
  Uxito-Ada

  authored
  fa261b8

• [NPU] further fix of new_value_states (#12538)

  rnwang04

  authored
  7cc01fd

• [NPU] Modify IPEX_LLM_NPU_DISABLE_COMPILE_OPT setting for long input (#12537)

  plusbang

  authored
  6596c18

• optimize glm edge again (#12539)

  MeouSker77

  authored
  1521994

• [NPU] Fix generate example (#12541)

  plusbang

  authored
  d20a968

• [Ollama] Update ipex-llm ollama readme to v0.4.6 (#12542)

  Show description for 5402fc6
  sgwhat

  authored
  5402fc6

• remove old rope usage (#12544)

  MeouSker77

  authored
  c090d16

• [NPU] Add IPEX_LLM_NPU_MTL to enable support on mtl (#12543)

  plusbang

  authored
  caf15cc

Commits on Dec 16, 2024

• [REFINE] graphmode code (#12540)

  ACupofAir

  authored
  0b953e6

• Add GLM-Edge GPU example (#12483)

  Show description for a86487c
  

  cranechu0131

  and

  Oscilloscope98

  authored
  a86487c

• remove old rope usage (#12552)

  MeouSker77

  authored
  5ae0006

• Add Modelscope option for chatglm3 on GPU (#12545)

  Show description for ccc18ee
  

  ATMxsp01

  and

  ATMxsp01

  authored
  ccc18ee

Commits on Dec 17, 2024

• [NPU doc] Update configuration for different platforms (#12554)

  plusbang

  authored
  680ea7e

• use new fused layer norm (#12553)

  MeouSker77

  authored
  a608f26

• Small typo fixes (#12558)

  Oscilloscope98

  authored
  d127a86

• [NPU] support asym_int4 for llama (#12556)

  Show description for fcb4748
  lzivan

  authored
  fcb4748

• Change: Use cn mirror for PyTorch extension installation to resolve network issues (#12559)

  Show description for 429bf1f
  liu-shaojun

  authored
  429bf1f

• [NPU doc] Add ARL runtime configuration (#12562)

  plusbang

  authored
  694d14b

• Add setuptools as a basic dependency (#12563)

  Show description for 6278caf
  Oscilloscope98

  authored
  6278caf

Commits on Dec 18, 2024

• Update readme (#12565)

  jason-dai

  authored
  6e801bc

• [NPU] support asym_int4 for minicpm (#12567)

  lzivan

  authored
  1a2ab12

• optimize siglip attention on arc (#12569)

  MeouSker77

  authored
  a4eb561

• small fix (#12573)

  MeouSker77

  authored
  e2ae429

• Update ollama and llama.cpp readme (#12574)

  sgwhat

  authored
  f7a2bd2

• Replace runner doc in ollama quickstart (#12575)

  sgwhat

  authored
  28e81fd

Commits on Dec 19, 2024

• Add --modelscope in GPU examples for glm4, codegeex2, qwen2 and qwen2.5 (#12561)

  Show description for 47e90a3
  

  ATMxsp01

  and

  ATMxsp01

  authored
  47e90a3

• padding mask on torch side (#12577)

  MeouSker77

  authored
  e0921f8

• optimize siglip attention again (#12578)

  MeouSker77

  authored
  4540424

• optimize new minicpm model (#12579)

  MeouSker77

  authored
  80f2fdc

• [NPU] Fix MTL and ARL support (#12580)

  plusbang

  authored
  4e7e988

• support Megrez-3B-Omni (#12582)

  MeouSker77

  authored
  3eeb02f

• Add --modelscope in GPU examples for minicpm, minicpm3, baichuan2 (#12564)

  Show description for 47da3c9
  

  ATMxsp01

  and

  ATMxsp01

  authored
  47da3c9

Commits on Dec 20, 2024

• Upgrade oneccl version to 0.0.6.3 (#12560)

  Show description for 51ff9eb
  liu-shaojun

  authored
  51ff9eb

• refactor qwen2 and llama3 (#12587)

  MeouSker77

  authored
  f3b5fad

• Add --modelscope option for glm-v4 MiniCPM-V-2_6 glm-edge and internvl2 (#12583)

  Show description for b0338c5
  

  ATMxsp01

  and

  ATMxsp01

  authored
  b0338c5

• refactor glm edge (#12588)

  MeouSker77

  authored
  6ea8033

• refactor yuan2 and starcoder2 and fix (#12589)

  MeouSker77

  authored
  b050368

• refactor sd 1.5 and qwen2-vl and fix (#12590)

  MeouSker77

  authored
  098eb33

Commits on Dec 24, 2024

• [NPU] support asym_int4 for baichuan (#12576)

  Show description for c410d9c
  lzivan

  authored
  c410d9c

• refactor baichuan, glm4 and minicpm3 (#12600)

  MeouSker77

  authored
  7aaf02f

• refactor mllama, gpt2 and internvl (#12602)

  MeouSker77

  authored
  ad2dc96

• [NPU] Fix minicpm on MTL (#12599)

  plusbang

  authored
  45f8f72

• refactor mistral and phi3 (#12605)

  MeouSker77

  authored
  073f936

• refactor chatglm2, internlm, stablelm and qwen (#12604)

  MeouSker77

  authored
  4135b89

• Update README.zh-CN.md (#12570)

  joan726

  authored
  9c9800b

Commits on Dec 25, 2024

• add compresskv back for mistral (#12607)

  Show description for 4e6b9d8
  MeouSker77

  authored
  4e6b9d8

• Update README.zh-CN.md (#12610)

  jason-dai

  authored
  54b1d7d

• fix llama related import (#12611)

  MeouSker77

  authored
  5f5ac8a

• rewrite llama optimization (#12609)

  MeouSker77

  authored
  6249c1e

Commits on Dec 26, 2024

• [docs] Update doc for latest open webui: 0.4.8 (#12591)

  Show description for 0477fe6
  Mingqi2

  authored
  0477fe6

• [NPU] fix npu save (#12614)

  Show description for 9e895f0
  rnwang04

  authored
  9e895f0

• remove bigdl-llm test to fix langchain UT (#12613)

  MeouSker77

  authored
  a596f1a

• Update Dockerfile (#12585)

  liu-shaojun

  authored
  28737c2

• Polish Readme for ModelScope-related examples (#12603)

  ATMxsp01

  authored
  ef585d3

• [NPU] update convert script based on latest usage (#12617)

  rnwang04

  authored
  d841e1d

• small fix (#12616)

  MeouSker77

  authored
  1604b4e

• [NPU] Update prompt format for baichuan2 (#12615)

  Show description for ccc4055
  lzivan

  authored
  ccc4055

• Consolidated C-Eval Benchmark Guide for Single-GPU and Multi-GPU Environments (#12618)

  Show description for 40a7d2b
  liu-shaojun

  authored
  40a7d2b

• support passing attn_scale to sdpa (#12619)

  MeouSker77

  authored
  a9abde0

• [NPU] Compatible with other third-party models like auto-round (#12620)

  Show description for bbdbbb0
  rnwang04

  authored
  bbdbbb0

• [NPU doc] Update verified platforms (#12621)

  plusbang

  authored
  796ee57