Pull requests: NVIDIA/TensorRT-LLM
Pull requests list
[None][fix] source DSA metadata from sparse params
#16236
opened Jul 10, 2026 by
liji-nv
Collaborator
Loading…
1 task done
[None][feat] GDS support, direct disk->gpu onboarding
#16235
opened Jul 10, 2026 by
reasonsolo
Collaborator
•
Draft
1 task
[TRTLLM-14026][feat] BREAKING: Remove TensorRT from the C++ tree
api-breaking
Accepted LLM API contract change that is backwards-incompatible
#16234
opened Jul 10, 2026 by
Wanli-Jiang
Collaborator
•
Draft
1 task done
[None][fix] Fix VSWA gate in KVCacheManager window-size resolution
#16233
opened Jul 10, 2026 by
eopXD
Collaborator
Loading…
1 task done
[None][perf] Memoize multimodal run metadata and drop identity block-offset gather
#16232
opened Jul 10, 2026 by
eopXD
Collaborator
Loading…
1 task done
[None][doc] Add tech blog: Evaluating Agentic Serving with Trace Replay and Job-Level Metrics
#16230
opened Jul 10, 2026 by
KleinBlueC
Contributor
•
Draft
1 task
[TRTLLM-12838][infra] enhance function-level code coverage for catching subprocess data
#16229
opened Jul 10, 2026 by
crazydemo
Collaborator
Loading…
1 task done
[TRTLLM-14234][perf] LTX-2: replicate audio stream across Ulysses ranks by default
#16228
opened Jul 10, 2026 by
luyiyun1021
Collaborator
Loading…
1 task done
[https://nvbugs/6400533][fix] Cache per-request greedy flag to fast-exit spec-dec sampling scan
#16227
opened Jul 10, 2026 by
trtllm-agent
Collaborator
Loading…
2 tasks done
[https://nvbugs/6426852][fix] Fix intermittent cuda mapping error
#16226
opened Jul 10, 2026 by
shuyixiong
Collaborator
Loading…
1 task done
[None][fix] Correct RocketKV KT cache byte accounting for FP8 KV cache
#16225
opened Jul 10, 2026 by
eopXD
Collaborator
Loading…
1 task done
[None][feat] Enable DeepSeek-V4 and DSA (DeepSeek-V3.2/GLM) serving on SM120 via FlashInfer sparse-MLA
#16224
opened Jul 10, 2026 by
lucifer1004
Collaborator
Loading…
[None][fix] Derive DSA indexer K-cache block stride at use time
#16223
opened Jul 10, 2026 by
lucifer1004
Collaborator
Loading…
[None][fix] clamp very small non-zero sampling temperature
#16222
opened Jul 10, 2026 by
lonexreb
Loading…
[None][fix] Fix fused mHC output reuse and extend compressor next_n
#16221
opened Jul 10, 2026 by
heyuhhh
Collaborator
Loading…
6 tasks done
[https://nvbugs/6422339][test] Unwaive 3 disagg-perf-sanity GB200 tests after #15961 revert
#16220
opened Jul 10, 2026 by
chenfeiz0326
Collaborator
Loading…
2 tasks done
[None][chore] KVCacheManagerV2: Python and test preparation for a C++ backend
api-compatible
Accepted LLM API contract change that is backwards-compatible
#16218
opened Jul 10, 2026 by
lowsfer
Member
Loading…
3 tasks done
[None][fix] Bump DeepGEMM to nv_dev HEAD for SM120 attention support
#16217
opened Jul 10, 2026 by
lucifer1004
Collaborator
Loading…
[None][fix] Propagate NCCL_INCLUDE_DIR to common header dirs
#16216
opened Jul 10, 2026 by
lucifer1004
Collaborator
Loading…
[None][fix] Generate CUTLASS kernel instantiations without installing cutlass_library
#16215
opened Jul 10, 2026 by
lucifer1004
Collaborator
Loading…
[TRTLLM-14022][feat] BREAKING: DON'T REVIEW Remove legacy TensorRT Python backend
api-breaking
Accepted LLM API contract change that is backwards-incompatible
#16213
opened Jul 10, 2026 by
Wanli-Jiang
Collaborator
•
Draft
1 task done
[https://nvbugs/6127669][fix] Fix layer-wise benchmarks performance alignment test
#16212
opened Jul 10, 2026 by
kaiyux
Member
Loading…
1 task done
[None][feat] Add fused T5 encoder self-attention path in bertAttentionPlugin
#16211
opened Jul 10, 2026 by
zhanghuanzj
Loading…
[None][test] Fix the multinode test case on DGX-Spark(perf skipped and func hang)
#16209
opened Jul 10, 2026 by
JennyLiu-nv
Collaborator
Loading…
1 task done
ProTip!
no:milestone will show everything without a milestone.