Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Jan 24, 2025

• [Model] Fused rope implementation for DeepSeek-v2 (#3105)

  Show description for 00f5303
  annanyapr

  authored
  00f5303

  Copy full SHA for 00f5303

Commits on Jan 22, 2025

• [Bench] Add AzureLLMInference dataset (#3104)

  Show description for 2c1001b
  MasterJH5574

  authored
  2c1001b

  Copy full SHA for 2c1001b

Commits on Jan 21, 2025

• [Bench] Add WildChat dataset (#3102)

  Show description for 03509ce
  MasterJH5574

  authored
  03509ce

  Copy full SHA for 03509ce

• [Docs] Revert the `--no-index` changes in instalation commands (#3101)

  Show description for c84e80a
  MasterJH5574

  authored
  c84e80a

  Copy full SHA for c84e80a

• [Fix] Fix deepseek r1 stop tokens (#3098)

  Show description for bd6cf37
  CharlieFRuan

  authored
  bd6cf37

  Copy full SHA for bd6cf37

Commits on Jan 20, 2025

• [Python] Add Deepseek-v3 conversation template (#3097)

  Show description for a175d44
  MasterJH5574

  authored
  a175d44

  Copy full SHA for a175d44

• [Fix] Add `--no-index` to pip install commands (#3096)

  Show description for b8838a1
  MasterJH5574

  authored
  b8838a1

  Copy full SHA for b8838a1

Commits on Jan 14, 2025

• [Format] pre-commit setup (#3090)

  Show description for 24d0796
  

  MasterJH5574

  and

  Ubospica

  authored
  24d0796

  Copy full SHA for 24d0796

• [Python] Add Microserving code example (#3089)

  Show description for 8845ae8
  jinhongyii

  authored
  8845ae8

  Copy full SHA for 8845ae8

Commits on Jan 8, 2025

• [Model] Reuse RoPE positions for Deepseek-v2 model (#3084)

  Show description for cf7ae82
  MasterJH5574

  authored
  cf7ae82

  Copy full SHA for cf7ae82

Commits on Jan 7, 2025

• Update copyright notice to 2023-2025 (#3085)

  Show description for bf70bea
  MasterJH5574

  authored
  bf70bea

  Copy full SHA for bf70bea

Commits on Jan 5, 2025

• [Model] Fix cached buffer allocation and fix dynamic shape in Phi3v (#3082)

  Show description for 6059763
  mengshyu

  authored
  6059763

  Copy full SHA for 6059763

Commits on Jan 4, 2025

• [Model] Support multi-GPU for Deepseek-v2 (#3080)

  Show description for 8243b2b
  MasterJH5574

  authored
  8243b2b

  Copy full SHA for 8243b2b

Commits on Jan 3, 2025

• [Compile] Reset custom allreduce flag in O2 mode to NONE (#3081)

  Show description for 6faf68e
  MasterJH5574

  authored
  6faf68e

  Copy full SHA for 6faf68e

Commits on Jan 2, 2025

• [Python] Update top-k kernel with TIR meta-programming (#3079)

  Show description for 3463ab7
  MasterJH5574

  authored
  3463ab7

  Copy full SHA for 3463ab7

Commits on Dec 19, 2024

• [SLM] GPTJ Multi-GPU support (#3070)

  Show description for 1825fed
  tlopex

  authored
  1825fed

  Copy full SHA for 1825fed

• [Model] Add support for Nemotron architecture (#3069)

  Show description for 9a33772
  hrishi121

  authored
  9a33772

  Copy full SHA for 9a33772

Commits on Dec 18, 2024

• [Serve] MicroServing API refactor (#3071)

  Show description for 8a1bfd6
  MasterJH5574

  authored
  8a1bfd6

  Copy full SHA for 8a1bfd6

Commits on Dec 17, 2024

• [Bench] Add "per-gpu-workload" mode (#3068)

  Show description for 88074ea
  MasterJH5574

  authored
  88074ea

  Copy full SHA for 88074ea

• [Model] Add support for GPTJ architecture (#3012)

  Show description for 88ebe6f
  tlopex

  authored
  88ebe6f

  Copy full SHA for 88ebe6f

Commits on Dec 16, 2024

• [Serve] MicroServing Implementation (#3064)

  Show description for 5c9ebcb
  

  4 peopleauthored
  5c9ebcb

  Copy full SHA for 5c9ebcb

• [CI] Update TVM package for CI (#3067)

  Show description for 49dcd4a
  MasterJH5574

  authored
  49dcd4a

  Copy full SHA for 49dcd4a

• Auto updated submodule references

  

  Git bot

  committed
  9dd0788

  Copy full SHA for 9dd0788

Commits on Dec 15, 2024

• Add prefix-cache dataset in mlc bench (#3065)

  Show description for e5f3880
  jinhongyii

  authored
  e5f3880

  Copy full SHA for e5f3880

Commits on Dec 14, 2024

• [Model] Add support for OLMo architecture (#3046)

  Show description for 385cef2
  Lanssi

  authored
  385cef2

  Copy full SHA for 385cef2

Commits on Dec 11, 2024

• [Python] Add flag to disable logging config (#3060)

  Show description for 86cf3f7
  hykilpikonna

  authored
  86cf3f7

  Copy full SHA for 86cf3f7

Commits on Dec 6, 2024

• [Fix][Model] Fix tensor parallelism of internlm2 model (#3058)

  Show description for 8477ac0
  MasterJH5574

  authored
  8477ac0

  Copy full SHA for 8477ac0

Commits on Nov 21, 2024

• [C++] Invoke storage allocation for CUDA Graph explicitly (#3042)

  Show description for d23d6f5
  MasterJH5574

  authored
  d23d6f5

  Copy full SHA for d23d6f5

• Update copyright to 2023-2024 (#3041)

  Show description for e349684
  MasterJH5574

  authored
  e349684

  Copy full SHA for e349684

• [Python] Skip termination when engine is not initialized (#3040)

  Show description for 77b7857
  MasterJH5574

  authored
  77b7857

  Copy full SHA for 77b7857

• Auto updated submodule references

  

  Git bot

  committed
  78b69f0

  Copy full SHA for 78b69f0

Commits on Nov 20, 2024

• Create NOTICE file

  tqchen

  authored
  53a6dcf

  Copy full SHA for 53a6dcf

Commits on Nov 16, 2024

• [JIT] Support overriding optimization flags in JIT (#3032)

  Show description for e283cd0
  MasterJH5574

  authored
  e283cd0

  Copy full SHA for e283cd0

• [Model] Keep vision encoder weights unquantized to maintain accuracy (#3028)

  Show description for 3578e79
  mengshyu

  authored
  3578e79

  Copy full SHA for 3578e79

Commits on Nov 15, 2024

• [Model] Update default prefill chunk size of Deepseek (#3009)

  Show description for 967fb76
  tlopex

  authored
  967fb76

  Copy full SHA for 967fb76

Pagination

Previous

Next