Yanbin JiangandYusheng Su 9fd6d0ec99 [LoRA] BF16 support + EP cuda-graph crash fix for experimental_sgl_trtllm MoE-LoRA (#28953) ...

Co-authored-by: Yusheng Su <yushengsu.thu@gmail.com>

2026-06-24 21:16:50 -07:00

..

backend

[LoRA] Fix chunked SGMV (csgmv) CUDA graph segment replay (#28371)

2026-06-16 19:19:07 -07:00

torch_ops

[Bug Fix] Fix broken sgemm_lora_a_graph_fwd due to invalid torch.mm() call (#24760)

2026-05-12 22:29:01 +03:00

triton_ops

[LoRA] Fix chunked SGMV (csgmv) CUDA graph segment replay (#28371)

2026-06-16 19:19:07 -07:00

trtllm_lora_temp

[LoRA] BF16 support + EP cuda-graph crash fix for experimental_sgl_trtllm MoE-LoRA (#28953)

2026-06-24 21:16:50 -07:00

deepseek_mla_correction.py

[lint] Enable Ruff UP037 to drop redundant quoted annotations (#27984)

2026-06-11 17:38:05 -07:00

eviction_policy.py

Feature/Fix multi lora scheduler blocking issue and evict LoRA None lastly (#14795)

2025-12-12 17:13:05 -08:00

layers.py

[LoRA] Experimental fast LoRA path with experimental_sgl_trtllm MoE backend for FP8 and NVFP4 models (#27329)

2026-06-05 14:45:59 -07:00

lora_config.py

throw ValueError for DoRA adapters (#22125)

2026-05-02 14:54:19 +00:00

lora_drainer.py

[Feature] add LoRADrainer to address high P99 TTFT (#17913)

2026-05-02 16:13:43 -07:00

lora_manager.py

[LoRA] Support DSA indexer LoRA targets for GLM-5.1 / DeepSeek-V3.2-family models (#28110)

2026-06-13 23:02:33 -07:00

lora_moe_runner_marlin.py

refactor(moe): de-duplicate triton MoE runner path into shared helpers (#23019)

2026-04-17 17:05:13 -07:00

lora_moe_runners.py

[Refactor] Cuda Graph Runner/Backend Refactor (#23906)

2026-06-09 21:36:57 -07:00

lora_overlap_loader.py

[lora] Fix overlap loading for cancelled requests (#25413)

2026-05-25 15:18:36 +09:00

lora_registry.py

[LoRA] Use deterministic lora_id for --lora-paths so multi-node ranks agree (#24555)

2026-05-06 22:20:15 -07:00

lora.py

[lora] More efficient pinned memory (#20876)

2026-05-30 09:04:59 +09:00

mem_pool.py

[LoRA] Experimental fast LoRA path with experimental_sgl_trtllm MoE backend for FP8 and NVFP4 models (#27329)

2026-06-05 14:45:59 -07:00

utils.py

[LoRA] Support DSA indexer LoRA targets for GLM-5.1 / DeepSeek-V3.2-family models (#28110)

2026-06-13 23:02:33 -07:00