Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Mar 13, 2025

• Adding a parameter=lora_input_adapters_path, to start the server with path if the path is available. Else None. (#223)

  Show description for a75f64f
  aman2930

  authored
  a75f64f

  Copy full SHA for a75f64f

Commits on Mar 11, 2025

• Add the llama2-70b model (#217)

  xy12181

  authored
  982569f

  Copy full SHA for 982569f

Commits on Mar 5, 2025

• Add unit tests to increase coverage for multi-sampling (#220)

  Show description for 2c4f6d9
  Lumosis

  authored
  2c4f6d9

  Copy full SHA for 2c4f6d9

• Chunked Prefill (#188)

  mailvijayasingh

  authored
  698f33f

  Copy full SHA for 698f33f

• Kill the generate thread when getting a "None" new request (#218)

  Show description for f602565
  Lumosis

  authored
  f602565

  Copy full SHA for f602565

• Revert accidental change - back to #216

  Show description for 1d6b456
  

  jyj0w0

  and

  Yijia J

  authored
  1d6b456

  Copy full SHA for 1d6b456

• Merge pull request #183 from AI-Hypercomputer:yijiaj/test

  Show description for 00dc5a6
  

  jetstream authors

  committed
  00dc5a6

  Copy full SHA for 00dc5a6

Commits on Mar 3, 2025

• Require at least 1GB of total HBM from all devices (#216)

  xy12181

  authored
  951b3ef

  Copy full SHA for 951b3ef

• Add multi-sampling functionality (#215)

  Lumosis

  authored
  80bfefc

  Copy full SHA for 80bfefc

Commits on Feb 28, 2025

• Update setup.py with latest release tag

  vipannalla

  authored
  84a67d1

  Copy full SHA for 84a67d1

• Add request id into active request (#209)

  liurupeng

  authored
  5a74bca

  Copy full SHA for 5a74bca

• [refactor] clean up model executor (#214)

  Show description for 66ee11a
  xy12181

  authored
  66ee11a

  Copy full SHA for 66ee11a

Commits on Feb 27, 2025

• [cleanup] Further clean up on prefill/inference/postprocess functions (#213)

  Show description for 9729700
  xy12181

  authored
  9729700

  Copy full SHA for 9729700

• [style] Use single underscore prefix for private data/functions (#212)

  Show description for 297a807
  xy12181

  authored
  297a807

  Copy full SHA for 297a807

• [cleanup] Refactor kv cache storage/manager (#211)

  Show description for 76b2d90
  xy12181

  authored
  76b2d90

  Copy full SHA for 76b2d90

Commits on Feb 26, 2025

• [cleanup] Refactor prefill functions (#210)

  Show description for 5dd2b91
  xy12181

  authored
  5dd2b91

  Copy full SHA for 5dd2b91

• [cleanup] Clean up logs (#208)

  xy12181

  authored
  345bcaf

  Copy full SHA for 345bcaf

Commits on Feb 25, 2025

• 1> Add mmlu dataset; 2> fix minor bug for mmlu prompt; 3> add scikit-… (#203)

  Show description for cddb367
  bzgoogle

  authored
  cddb367

  Copy full SHA for cddb367

• Fix the Math benchmark issues. (#207)

  Show description for 577e0e2
  wang2yn84

  authored
  577e0e2

  Copy full SHA for 577e0e2

• [cleanup] Move all inference config parameters into one file (#204)

  xy12181

  authored
  4f50332

  Copy full SHA for 4f50332

• Fix the benchmark_serving minor error. (#202)

  wang2yn84

  authored
  947a947

  Copy full SHA for 947a947

Commits on Feb 24, 2025

• [test] Move unit tests to the same folder as their source code (#200)

  Show description for 85bcc3a
  xy12181

  authored
  85bcc3a

  Copy full SHA for 85bcc3a

Commits on Feb 21, 2025

• [cleanup/fix] Make inference parameters explicit, etc. (#198)

  Show description for ec77720
  xy12181

  authored
  ec77720

  Copy full SHA for ec77720

• [Deepseek] enable mmlu benchmark (#193)

  Show description for 85064bd
  bzgoogle

  authored
  85064bd

  Copy full SHA for 85064bd

Commits on Feb 20, 2025

• Add Math-500 dataset from Huggingface. (#199)

  Show description for 01fcc2a
  wang2yn84

  authored
  01fcc2a

  Copy full SHA for 01fcc2a

Commits on Feb 19, 2025

• [benchmark] Increase offline benchmarking batch size from 160 to 320 (#196)

  Show description for 8cae8a1
  xy12181

  authored
  8cae8a1

  Copy full SHA for 8cae8a1

• [test] Add a unit test for paged attention kernel (#195)

  Show description for 7f831bc
  xy12181

  authored
  7f831bc

  Copy full SHA for 7f831bc

• [test] fix unit test breaks (#194)

  xy12181

  authored
  88a2af4

  Copy full SHA for 88a2af4

Commits on Feb 17, 2025

• Add comprehensive debugging messages. (#182)

  Show description for 2527fe4
  Lumosis

  authored
  2527fe4

  Copy full SHA for 2527fe4

• [MLPerf Doc and Script] Refresh readme and script. Reduce logging in benchmarks/mlperf/backend.py (#190)

  Show description for a2ddb48
  wyzhang

  authored
  a2ddb48

  Copy full SHA for a2ddb48

Commits on Feb 16, 2025

• [TestCleanup] Clean up mock_engine and its test (#189)

  Show description for 0f9e688
  

  wyzhang

  and

  vipannalla

  authored
  0f9e688

  Copy full SHA for 0f9e688

Commits on Feb 15, 2025

• [DocUpdate] Add jax installation on TPU VM step to mlperf instruction

  Show description for e9c4083
  wyzhang

  committed
  e9c4083

  Copy full SHA for e9c4083

Commits on Feb 13, 2025

• More chagnes to mlperf setup to support llama2 70b and mixtral 8x7b (#187)

  vipannalla

  authored
  101545e

  Copy full SHA for 101545e

• [TestCleanup] Clean up test_server.py (#186)

  Show description for 8f91427
  wyzhang

  authored
  8f91427

  Copy full SHA for 8f91427

Commits on Feb 10, 2025

• More chagnes to mlperf setup to support llama2 70b and mixtral 8x7b (#185)

  vipannalla

  authored
  cd44c56

  Copy full SHA for cd44c56

Pagination

Previous

Next