TensorRT-LLMs/tensorrt_llm/_torch
Netanel Haber d9a3530048
[nvbug/5393888][nvbug/5393042] Always use py_seq_slot (#6147)
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
2025-07-18 22:45:16 +03:00
..
attention_backend feat: TRTLLM-5574 Add phi-4-multimodal pytorch-backend support (#5644) 2025-07-17 06:30:58 +08:00
auto_deploy [refactor] Simplification of Speculative decoding configs (#5639) 2025-07-10 11:37:30 -04:00
compilation [feat] Piecewise cuda graph support for MLA (#4467) 2025-06-17 18:58:38 +08:00
custom_ops [TRTLLM-6100] fix: Nvbug 5356427: autotuned TRTLLM Gen fp8 block scale MoE illegal memory access (#5676) 2025-07-14 17:17:30 +08:00
debug Add debug hook to support dump tensor data and add new debug functions easily (#5182) 2025-06-24 17:45:28 +08:00
distributed [feat] Support torch compile for attention dp (#5086) 2025-07-01 13:48:52 -04:00
models feat(eagle3):support qwen3 dense model (#5879) 2025-07-19 01:24:32 +08:00
modules feat: Remove padding in attention DP. (#6064) 2025-07-18 23:30:34 +08:00
peft feat: support multi lora adapters and TP (#3885) 2025-05-08 23:45:45 +08:00
pyexecutor [nvbug/5393888][nvbug/5393042] Always use py_seq_slot (#6147) 2025-07-18 22:45:16 +03:00
shared_tensor [1/N][TRTLLM-5195][feat] Share PyTorch tensor between processes (#5396) 2025-07-10 05:12:53 +09:00
speculative [nvbug/5393888][nvbug/5393042] Always use py_seq_slot (#6147) 2025-07-18 22:45:16 +03:00
__init__.py [TRTLLM-5493] Add core infrastructure to enable loading of custom checkpoint formats (#5372) 2025-07-17 00:50:30 +08:00
autotuner.py [TRTLLM-5770] feat: Integrate TRT-LLM Gen FP8 block scale MoE with Pytorch workflow kernel autotuner (#5207) 2025-06-17 21:01:56 +08:00
expert_statistic.py Add MTP support for Online EPLB (#5213) 2025-06-25 07:58:13 +08:00
llm.py [TRTLLM-5208][BREAKING CHANGE] chore: make pytorch LLM the default (#5312) 2025-06-20 03:01:10 +08:00
metadata.py feat: no-cache attention in PyTorch workflow (#3085) 2025-04-05 01:54:32 +08:00
model_config.py feat: add support for Modelopt fp8_pb_wo quantization scheme (#6106) 2025-07-18 10:35:12 +08:00
utils.py chore: Cleanup disable_fp4_allgather. (#6006) 2025-07-16 17:54:36 +08:00