kanshan/TensorRT-LLMs
1
0
Fork 0
mirror of https://github.com/NVIDIA/TensorRT-LLM.git synced 2026-01-14 06:27:45 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity
b8818b45be
TensorRT-LLMs/tensorrt_llm/_torch/models
History
Chang Liu b8818b45be
fix: llama4: address couple of issues in llama4 attention module (#3491) 
* fix attn module for llama4

* Address comments

* Rebase to accommodate latest attn refactor and refactor l4attn

* Remove aux_stream from classic attn

* Use RMSNorm for L2Norm

* Update tensorrt_llm/_torch/models/modeling_llama.py

Co-authored-by: hlu1 <14827759+hlu1@users.noreply.github.com>
Signed-off-by: Chang Liu <lc9114@gmail.com>

* Add typing informations for _attn_qkv

* Remove redundant comment

* Simplify llama4 DecoderLayer logic

---------

Signed-off-by: Chang Liu <lc9114@gmail.com>
Co-authored-by: hlu1 <14827759+hlu1@users.noreply.github.com>
2025-04-18 01:54:59 +00:00
..
__init__.py feat: Nemotron-H model support (#3430) 2025-04-16 14:05:56 -07:00
.gitkeep Update TensorRT-LLM (#2755) 2025-02-11 03:01:00 +00:00
modeling_auto.py Cache sin cos in model instead of global LRU cache. (#3378) 2025-04-14 11:19:09 +08:00
modeling_bert.py feat: Support cos_sin_cache in all cases. (#3517) 2025-04-16 13:48:44 +08:00
modeling_deepseekv3.py Clean up linear.py, mlp.py, gated_mlp.py (#3553) 2025-04-16 12:21:44 -07:00
modeling_llama.py fix: llama4: address couple of issues in llama4 attention module (#3491) 2025-04-18 01:54:59 +00:00
modeling_llava_next.py Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
modeling_mamba_hybrid.py feat: Support cos_sin_cache in all cases. (#3517) 2025-04-16 13:48:44 +08:00
modeling_mixtral.py feat: Support cos_sin_cache in all cases. (#3517) 2025-04-16 13:48:44 +08:00
modeling_mllama.py chore: Use ellipsis as default value to detect whether residual argument is provided (#3626) 2025-04-17 12:31:58 +08:00
modeling_multimodal_encoder.py Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
modeling_multimodal_utils.py Update TensorRT-LLM (#2873) 2025-03-11 21:13:42 +08:00
modeling_nemotron_h.py Fix rotary_emb param in NemotronH attention (#3646) 2025-04-16 21:03:07 -07:00
modeling_nemotron_nas.py feat: Support cos_sin_cache in all cases. (#3517) 2025-04-16 13:48:44 +08:00
modeling_nemotron.py feat: Support cos_sin_cache in all cases. (#3517) 2025-04-16 13:48:44 +08:00
modeling_qwen2vl.py feat: adding multimodal (only image for now) support in trtllm-bench (#3490) 2025-04-18 07:06:16 +08:00
modeling_qwen_moe.py feat: Support cos_sin_cache in all cases. (#3517) 2025-04-16 13:48:44 +08:00
modeling_qwen.py feat: Support cos_sin_cache in all cases. (#3517) 2025-04-16 13:48:44 +08:00
modeling_utils.py chore: Use ellipsis as default value to detect whether residual argument is provided (#3626) 2025-04-17 12:31:58 +08:00
modeling_vila.py Add Llama 4 (#3302) 2025-04-09 03:35:21 +08:00