Commits

Branch selector

 main

User selector

pavanimajety

All time

Commit History

Commits on Feb 10, 2026

• [SM100] Resubmit FMHA FP8 prefill for MLA (#31195)

  Show description for 578977b
  pavanimajety

  authored
  578977b

  Copy full SHA for 578977b

Commits on Feb 5, 2026

• [Bugfix] Kimi-K2 grouped_topk usage for Flashinfer monolithic kernels. (#33858)

  Show description for d2f4a71
  pavanimajety

  authored
  d2f4a71

  Copy full SHA for d2f4a71

Commits on Jan 30, 2026

• [Hardware][SM100] Add TRTLLM Kernel for INT4 W4A16 Kernel. (#32437)

  Show description for c3a9752
  pavanimajety

  authored
  c3a9752

  Copy full SHA for c3a9752

Commits on Dec 23, 2025

• Revert "[SM100] Enable fp8 compute for prefill MLA (#30746)" (#31197)

  Show description for 3e10262
  pavanimajety

  authored
  3e10262

  Copy full SHA for 3e10262

Commits on Dec 22, 2025

• [SM100] Enable fp8 compute for prefill MLA (#30746)

  Show description for b10f41c
  pavanimajety

  authored
  b10f41c

  Copy full SHA for b10f41c

Commits on Nov 7, 2025

• [Bugfix] Use latency MOE backend as default for Flashinfer and other misc fixes (#27439)

  Show description for 72b1c2a
  pavanimajety

  authored
  72b1c2a

  Copy full SHA for 72b1c2a

Commits on Oct 21, 2025

• Add @pavanimajety to .github/codeowners for Flashinfer, ModelOpt related code (#27213)

  Show description for ecc3c09
  pavanimajety

  authored
  ecc3c09

  Copy full SHA for ecc3c09

Commits on Oct 3, 2025

• [Quantization/NVFP4] Speed up TRTLLM NVFP4 MOE weight loading and fix K/V scale loading for MLA Attn (#25968)

  Show description for a269173
  pavanimajety

  authored
  a269173

  Copy full SHA for a269173

Commits on Sep 30, 2025

• [Bugfix] Fix accuracy issue of TRTLLM FP8 MOE and improve logging (#25895)

  Show description for ef28354
  pavanimajety

  authored
  ef28354

  Copy full SHA for ef28354

Commits on Aug 21, 2025

• [Core] Always use tensor cores for Flashinfer Decode Wrapper (#23214)

  Show description for 1d353b6
  pavanimajety

  authored
  1d353b6

  Copy full SHA for 1d353b6

Commits on Jul 15, 2025

• [Bugfix] Switch bailout logic for kv-cache-dtype with SM100 Flashinfer (#20934)

  Show description for 9ad0a45
  pavanimajety

  authored
  9ad0a45

  Copy full SHA for 9ad0a45

Commits on Jul 11, 2025

• [Core] Add Flashinfer TRTLLM Backend for Flashinfer decode path (SM100). (#19825)

  Show description for 7bd4c37
  

  3 peopleauthored
  7bd4c37

  Copy full SHA for 7bd4c37

Commits on Jun 9, 2025

• [HOT-FIX] Add `kv_sharing_target_layer_name` argument to cutlass_mla backend (#19374)

  Show description for 8058c91
  pavanimajety

  authored
  8058c91

  Copy full SHA for 8058c91

Commits on May 23, 2025

• [ModelOpt] Introduce VLLM_MAX_TOKENS_PER_EXPERT_FP4_MOE env var to control blockscale tensor allocation (#18160)

  Show description for f203673
  pavanimajety

  authored
  f203673

  Copy full SHA for f203673

Commits on May 14, 2025

• [Hardware/NVIDIA/Modelopt] Fix modelopt forward method for v1 torch.compile (#18101)

  Show description for 65f0f74
  pavanimajety

  authored
  65f0f74

  Copy full SHA for 65f0f74

Commits on May 9, 2025

• [Hardware/NVIDIA/Kernel] Enable nvidia/DeepSeek-R1-FP4 Model (#16362)

  pavanimajety

  authored
  0c0fdae

  Copy full SHA for 0c0fdae

Commits on Mar 12, 2025

• [Kernel] Add ModelOpt FP4 Checkpoint Support (#12520)

  Show description for debd6bb
  pavanimajety

  authored
  debd6bb

  Copy full SHA for debd6bb

Commits on Mar 6, 2025

• [Hardware] Update the flash attn tag to support Blackwell (#14244)

  pavanimajety

  authored
  ed6ea06

  Copy full SHA for ed6ea06

Commits on Jan 29, 2025

• [Hardware][NV] Fix Modelopt model loading for k-v-scales for Llama models. (#11787)

  Show description for b02fd28
  

  pavanimajety

  and

  mgoin

  authored
  b02fd28

  Copy full SHA for b02fd28

Commits on Nov 21, 2024

• [Core] Add Sliding Window Support with Flashinfer (#10462)

  Show description for 6c1208d
  pavanimajety

  authored
  6c1208d

  Copy full SHA for 6c1208d

Commits on Nov 13, 2024

• [Core] Flashinfer - Remove advance step size restriction (#10282)

  pavanimajety

  authored
  b6dde33

  Copy full SHA for b6dde33

Commits on Nov 1, 2024

• [Bugfix/Core] Flashinfer k_scale and v_scale (#9861)

  pavanimajety

  authored
  598b6d7

  Copy full SHA for 598b6d7

Commits on Sep 11, 2024

• [Hardware][NV] Add support for ModelOpt static scaling checkpoints. (#6112)

  pavanimajety

  authored
  efcf946

  Copy full SHA for efcf946

Commits on Aug 31, 2024

• [Bugfix] bugfix and add model test for flashinfer fp8 kv cache. (#8013)

  pavanimajety

  authored
  622f8ab

  Copy full SHA for 622f8ab

Commits on Aug 29, 2024

• [Core][Kernels] Enable FP8 KV Cache with Flashinfer backend. + BugFix for kv_cache_dtype=auto (#7985)

  Show description for 6b34215
  

  3 peopleauthored
  6b34215

  Copy full SHA for 6b34215

Commits on Aug 28, 2024

• [Core][Kernels] Use FlashInfer backend for FP8 KV Cache when available. (#7798)

  Show description for b98cc28
  

  pavanimajety

  and

  simon-mo

  authored
  b98cc28

  Copy full SHA for b98cc28