William Zhang c0ae6bbdbe [None][feat] EPD for Qwen3 VL (#10470) ... * Why? We would like to support EPD disaggregated serving for Qwen3 VL. * What? This commit adds such support, and extends existing unit tests for correctness checks. Some minor (protected) interface changes had to be made to the weight mapper as a side-effect. Signed-off-by: William Zhang <133824995+2ez4bz@users.noreply.github.com>		2026-01-08 06:45:54 -05:00
..
attention	[TRTLLM-9798][feat] Change to use new DeepGEMM MQA sm100 kernel for MTP-3 (#10226)	2025-12-24 14:39:12 +08:00
auto_deploy	[https://nvbugs/5747878][fix] unwaive llama4 scout tests (#10468)	2026-01-07 23:33:45 -05:00
compilation	[TRTLLM-3105][feat] Add Piecewise CUDA Graph Support (#3804)	2025-05-09 11:04:01 +08:00
debugger	Fix: fix nvbug 5356427 (#5464)	2025-06-25 22:24:26 +08:00
distributed	[TRTLLM-9467][fix] Fix PP+CP combination with helix parallelism (#10312)	2026-01-01 13:42:53 -05:00
executor	[https://nvbugs/5717993][fix] Add execution_stream across PyExecutor, KVCacheManager, PeftCacheManager to ensure proper CUDA stream synchronization between KV cache transfer operations and model forward kernels. (#10060)	2025-12-31 09:22:54 -08:00
misc	[None][perf] TRTLLM MoE maps to lower tuning buckets when ep>1 (#9998)	2026-01-05 17:16:12 +01:00
modeling	[None][feat] support Qwen3-VL dense model in pytorch backend (#9060)	2025-12-31 17:54:26 +09:00
models/checkpoints/hf	[TRTLLM-7136][feat] Update load_weights method to include mapping parameter in checkpoint loaders (#9583)	2025-12-05 16:07:20 +01:00
modules	[https://nvbugs/5784543][fix] Setup dist before using autotuner. (#10491)	2026-01-08 10:32:50 +08:00
multi_gpu	[TRTLLM-8821][feat] Apply AutoTuner to AllReduce Op for strategy tuning. (#8531)	2026-01-05 15:44:37 +08:00
multi_gpu_modeling	[https://nvbugs/5515753][ci] Add NCCL_DEBUG=INFO flag to collect more info with CI failure. (#8440)	2025-11-20 12:43:13 -05:00
multimodal	[None][feat] EPD for Qwen3 VL (#10470)	2026-01-08 06:45:54 -05:00
ray_orchestrator	[TRTLLM-9467][fix] Fix PP+CP combination with helix parallelism (#10312)	2026-01-01 13:42:53 -05:00
sampler	[None][fix] avoid implicit cudaStreamSynchronize in sample_async. (#10120)	2025-12-23 10:15:40 +08:00
speculative	[https://nvbugs/5749988][fix] Remove redundant qwen3 spec dec test (#10387)	2026-01-06 11:46:34 -05:00
thop	[None][feat] CuteDSL MOE FC1 Enhancement (#10088)	2026-01-06 09:30:43 +08:00
helpers.py	[#8733][feat] Add Llama4 MoE handling to AutoDeploy (#9556)	2025-12-04 08:03:33 +02:00
pattern_watcher.py	[TRTLLM-3105][feat] Add Piecewise CUDA Graph Support (#3804)	2025-05-09 11:04:01 +08:00
test_connector.py	[None][feat] KV Cache Connector API (#7228)	2025-08-28 23:09:27 -04:00
test_model_config.py	[TRTLLM-10171][fix] Correct attention handling in ModelConfig and KVCacheManager (#10330)	2026-01-04 06:07:30 -05:00