Yueh-Ting (eop) Chen a38d91aae2 [https://nvbugs/5537996][fix] Let KV cache manager block initialization be aware whether it is doing a dry run or not (#9093) ... Before this commit, the kv cache manager does the same regardless, which causes a mis-calculation in free memory available to allocate for the KV cache manager, hence causing a crash. This commit fixes this by letting KV cache manager initialization be aware whether it is doing the dry run or not. If it is a dry run, use the max_tokens setting that is already pre-calculated and filled into kv_cache_config.max_tokens. Signed-off-by: eopXD <yuehtingc@nvidia.com>		2025-11-25 17:27:11 +08:00
..
attention_backend	[https://nvbugs/5676748][fix] Fix mismatched nvfp4 gemm sf shape. (#9336)	2025-11-24 12:16:32 +08:00
auto_deploy	[#9413][fix] Minor fixes to nemotron H and custom models in AD (#9416)	2025-11-24 20:17:33 -08:00
compilation	[https://nvbugs/5546510][fix] Move torch.cuda.Stream out of torch com… (#8494)	2025-11-20 12:43:13 -05:00
configs	[TRTLLM-8535][feat] Support DeepSeek V3.2 with FP8 + BF16 KV cache/NVFP4 + BF16 KV cache (#8405)	2025-10-24 13:40:41 -04:00
custom_ops	[TRTLLM-7963][feat] Cold L2 cache when doing autotune benchmarking. (#8779)	2025-11-25 15:06:22 +08:00
cute_dsl_kernels	[TRTLLM-9370][feat] Integration of CuteDSL NVFP4 grouped GEMM (Part 2: SwiGLU Fusion and Finalize Fusion) (#9288)	2025-11-21 14:03:38 -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	[None][chore] Reduce nested nvtx ranges. (#9347)	2025-11-25 09:58:41 +08:00
models	[None][feat] Support Yarn on QwQ-32B model (#9059)	2025-11-25 07:27:28 +08:00
modules	[None][feat] Support Yarn on QwQ-32B model (#9059)	2025-11-25 07:27:28 +08:00
peft	[TRTLLM-7346][fix] Improve performance of PyTorchModelEngine._get_lora_params_from_requests (#7033)	2025-08-25 10:37:40 +03:00
pyexecutor	[https://nvbugs/5537996][fix] Let KV cache manager block initialization be aware whether it is doing a dry run or not (#9093)	2025-11-25 17:27:11 +08:00
shared_tensor	[1/N][TRTLLM-5195][feat] Share PyTorch tensor between processes (#5396)	2025-07-10 05:12:53 +09:00
speculative	[https://nvbugs/5590408][fix] Fallback to greedy sampling in two-model overlap scheduler (#9321)	2025-11-21 10:19:59 -05:00
__init__.py	[TRTLLM-9212][chore] move MoeLoadBalancerConfig to llm_args.py (#9002)	2025-11-13 10:47:35 +08:00
autotuner.py	[TRTLLM-7963][feat] Cold L2 cache when doing autotune benchmarking. (#8779)	2025-11-25 15:06:22 +08:00
cublaslt_utils.py	[https://nvbugs/5451205][feat] Add cuBLASLt NVFP4 GEMM backend support (#7943)	2025-10-23 15:55:10 +08:00
cute_dsl_utils.py	[None][chore] polish error message in cute_dsl_utils.py (#7852)	2025-09-19 12:05:11 +08:00
device_mesh.py	[TRTLLM-8682][chore] Remove auto_parallel module (#8329)	2025-10-22 20:53:08 -04:00
expert_statistic.py	[TRTLLM-8831][feat] Enable early exit with overlap scheduler (#8587)	2025-11-17 18:07:13 +01:00
flashinfer_utils.py	[None][ci] move unittests to sub-directories (#6635)	2025-08-20 05:42:22 -04:00
hostfunc.py	[TRTLLM-7028][feat] Enable guided decoding with speculative decoding (part 2: one-model engine) (#6948)	2025-09-03 15:16:11 -07:00
llm.py	[TRTLLM-5208][BREAKING CHANGE] chore: make pytorch LLM the default (#5312)	2025-06-20 03:01:10 +08:00
memory_buffer_utils.py	[https://nvbugs/5629833][fix] Don't fill tensors with 0 (#9296)	2025-11-21 20:50:05 +08:00
metadata.py	[None][feat] Use Separate QKV Input Layout for Context MLA (#6538)	2025-08-19 22:04:48 +08:00
model_config.py	[None][fix] Update the attention layers counting for Qwen3-next. (#9072)	2025-11-16 11:52:56 -08:00
utils.py	[https://nvbugs/5667687][fix] Set correct lm_head_tp_size_upper_bound (#9300)	2025-11-20 00:41:00 -08:00
virtual_memory.py	[TRTLLM-8511][feat] Add update_weights and sleep_wakeup support for rl integration (#8302)	2025-11-04 10:19:24 -08:00