Commits

Branch selector

 main

User selector

mgoin

Datepicker

All time

Commit History

Commits on Aug 28, 2024

• Remove faulty Meta-Llama-3-8B-Instruct-FP8.yaml lm-eval test (vllm-project#7961)

  mgoin

  authored
  af59df0

  Copy full SHA for af59df0

Commits on Aug 22, 2024

• [Bugfix] Use LoadFormat values for vllm serve --load-format (vllm-project#7784)

  mgoin

  authored
  15310b5

  Copy full SHA for 15310b5

• Revert "[Kernel] Expand MoE weight loading + Add Fused Marlin MoE Kernel (vllm-project#7527)" (vllm-project#7764)

  mgoin

  authored
  aae74ef

  Copy full SHA for aae74ef

Commits on Aug 16, 2024

• [Model] Align nemotron config with final HF state and fix lm-eval-small (vllm-project#7611)

  mgoin

  authored
  44f26a9

  Copy full SHA for 44f26a9

• [Doc] Update quantization supported hardware table (vllm-project#7595)

  mgoin

  authored
  d4f0f17

  Copy full SHA for d4f0f17

• [Doc] Add docs for llmcompressor INT8 and FP8 checkpoints (vllm-project#7444)

  mgoin

  authored
  b3f4e17

  Copy full SHA for b3f4e17

• [Kernel] Add tuned triton configs for ExpertsInt8 (vllm-project#7601)

  mgoin

  authored
  855866c

  Copy full SHA for 855866c

• [CI] Move quantization cpu offload tests out of fastcheck (vllm-project#7574)

  mgoin

  authored
  e165528

  Copy full SHA for e165528

• [Bugfix][Harmless] Fix float16 dtype for model_is_embedding (vllm-project#7566)

  mgoin

  authored
  9c8e2d1

  Copy full SHA for 9c8e2d1

Commits on Aug 15, 2024

• [Bugfix] Fix default weight loading for scalars (vllm-project#7534)

  mgoin

  authored
  21313e0

  Copy full SHA for 21313e0

Commits on Aug 8, 2024

• [Doc] Put collect_env issue output in a <detail> block (vllm-project#7310)

  mgoin

  authored
  e14fb22

  Copy full SHA for e14fb22

Commits on Aug 7, 2024

• [Doc] Update supported_hardware.rst (vllm-project#7276)

  mgoin

  authored
  6d94420

  Copy full SHA for 6d94420

• [Bugfix][FP8] Fix dynamic FP8 Marlin quantization (vllm-project#7219)

  mgoin

  authored
  5223199

  Copy full SHA for 5223199

• [Bugfix] Fix GPTQ and GPTQ Marlin CPU Offloading (vllm-project#7225)

  mgoin

  authored
  f9a5600

  Copy full SHA for f9a5600

Commits on Aug 2, 2024

• [CI/Build] Add support for Python 3.12 (vllm-project#7035)

  mgoin

  authored
  b482b9a

  Copy full SHA for b482b9a

Commits on Aug 1, 2024

• [Bugfix] Lower gemma's unloaded_params exception to warning (vllm-project#7002)

  mgoin

  authored
  f4fd390

  Copy full SHA for f4fd390

• [CI/Build] Remove sparseml requirement from testing (vllm-project#7037)

  mgoin

  authored
  fb3db61

  Copy full SHA for fb3db61

Commits on Jul 31, 2024

• [Misc] Add compressed-tensors to optimized quant list (vllm-project#7006)

  mgoin

  authored
  a0dce93

  Copy full SHA for a0dce93

• [Bugfix] Support cpu offloading with fp8 quantization (vllm-project#6960)

  mgoin

  authored
  460c188

  Copy full SHA for 460c188

Commits on Jul 27, 2024

• Add Nemotron to PP_SUPPORTED_MODELS (vllm-project#6863)

  mgoin

  authored
  b1366a9

  Copy full SHA for b1366a9

Commits on Jul 26, 2024

• [Doc] Add Nemotron to supported model docs (vllm-project#6843)

  mgoin

  authored
  281977b

  Copy full SHA for 281977b

• [Model] Support Nemotron models (Nemotron-3, Nemotron-4, Minitron) (vllm-project#6611)

  mgoin

  authored
  07278c3

  Copy full SHA for 07278c3

Commits on Jul 25, 2024

• [Bugfix] Fix kv_cache_dtype=fp8 without scales for FP8 checkpoints (vllm-project#6761)

  mgoin

  authored
  65b1f12

  Copy full SHA for 65b1f12

Commits on Jul 24, 2024

• [Bugfix] Bump transformers to 4.43.2 (vllm-project#6752)

  mgoin

  authored
  421e218

  Copy full SHA for 421e218

Commits on Jul 23, 2024

• [CI] Add smoke test for non-uniform AutoFP8 quantization (vllm-project#6702)

  mgoin

  authored
  01c16ed

  Copy full SHA for 01c16ed

• [Misc] Add ignored layers for fp8 quantization (vllm-project#6657)

  mgoin

  authored
  0eb0757

  Copy full SHA for 0eb0757

• [Misc] Support FP8 kv cache scales from compressed-tensors (vllm-project#6528)

  mgoin

  authored
  9e0b558

  Copy full SHA for 9e0b558

Commits on Jul 20, 2024

• [Misc] Fix input_scale typing in w8a8_utils.py (vllm-project#6579)

  mgoin

  authored
  f952bbc

  Copy full SHA for f952bbc

Commits on Jul 18, 2024

• [Model] Support Mistral-Nemo (vllm-project#6548)

  mgoin

  authored
  15c6a07

  Copy full SHA for 15c6a07

Commits on Jul 16, 2024

• [Kernel][Attention] Separate Attention.kv_scale into k_scale and v_scale (vllm-project#6081)

  mgoin

  authored
  978aed5

  Copy full SHA for 978aed5

Commits on Jul 12, 2024

• [Misc] Add generated git commit hash as vllm.__commit__ (vllm-project#6386)

  mgoin

  authored
  111fc6e

  Copy full SHA for 111fc6e

• [Model][Phi3-Small] Remove scipy from blocksparse_attention (vllm-project#6343)

  mgoin

  authored
  d59eb98

  Copy full SHA for d59eb98

Commits on Jul 3, 2024

• [Kernel] Expand FP8 support to Ampere GPUs using FP8 Marlin (vllm-project#5975)

  mgoin

  authored
  47f0954

  Copy full SHA for 47f0954

Commits on Jun 28, 2024

• [Bugfix] Only add Attention.kv_scale if kv cache quantization is enabled (vllm-project#5936)

  mgoin

  authored
  4bf35ed

  Copy full SHA for 4bf35ed

Commits on Jun 25, 2024

• [CI/Build] Add unit testing for FlexibleArgumentParser (vllm-project#5798)

  mgoin

  authored
  d9b34ba

  Copy full SHA for d9b34ba

Pagination

Previous

Next