TensorRT-LLM v0.18.2 release (#3611)

kaiyux · web-flow · commit 5aec7af45fc0 · 2025-04-16T14:42:50.000+08:00
diff --git a/README.md b/README.md
@@ -9,7 +9,7 @@ TensorRT-LLM
 [![python](https://img.shields.io/badge/python-3.10-green)](https://www.python.org/downloads/release/python-31012/)
 [![cuda](https://img.shields.io/badge/cuda-12.8.1-green)](https://developer.nvidia.com/cuda-downloads)
 [![trt](https://img.shields.io/badge/TRT-10.9.0-green)](https://developer.nvidia.com/tensorrt)
-[![version](https://img.shields.io/badge/release-0.18.1-green)](./tensorrt_llm/version.py)
+[![version](https://img.shields.io/badge/release-0.18.2-green)](./tensorrt_llm/version.py)
 [![license](https://img.shields.io/badge/license-Apache%202-blue)](./LICENSE)
 
 [Architecture](./docs/source/architecture/overview.md)&nbsp;&nbsp;&nbsp;|&nbsp;&nbsp;&nbsp;[Performance](./docs/source/performance/perf-overview.md)&nbsp;&nbsp;&nbsp;|&nbsp;&nbsp;&nbsp;[Examples](./examples/)&nbsp;&nbsp;&nbsp;|&nbsp;&nbsp;&nbsp;[Documentation](./docs/source/)&nbsp;&nbsp;&nbsp;|&nbsp;&nbsp;&nbsp;[Roadmap](https://docs.google.com/presentation/d/1gycPmtdh7uUcH6laOvW65Dbp9F1McUkGDIcAyjicBZs/edit?usp=sharing)
diff --git a/cpp/tensorrt_llm/batch_manager/aarch64-linux-gnu/version.txt b/cpp/tensorrt_llm/batch_manager/aarch64-linux-gnu/version.txt
@@ -1,2 +1,2 @@
 9f9942768fd5b0cf5ed19860ad539dc9 libtensorrt_llm_ucx_wrapper.so
-d2efc6043262c896e262e8d8b97055af0f1f8b47 commit
+edf502396e4443f284a5fae6044402478cf457c1 commit
diff --git a/cpp/tensorrt_llm/batch_manager/x86_64-linux-gnu/libtensorrt_llm_batch_manager_static.a b/cpp/tensorrt_llm/batch_manager/x86_64-linux-gnu/libtensorrt_llm_batch_manager_static.a
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e015c5cab637202b76f6ccd2d59b1427dc739f10321996a003230ba32814c08b
+oid sha256:ee2a324ae76a843823d1d82686bb495d097367e1c3a41aa9596fd0d2ba3fadae
 size 8408224
diff --git a/cpp/tensorrt_llm/batch_manager/x86_64-linux-gnu/libtensorrt_llm_batch_manager_static.pre_cxx11.a b/cpp/tensorrt_llm/batch_manager/x86_64-linux-gnu/libtensorrt_llm_batch_manager_static.pre_cxx11.a
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1205f8fe60cc657645441c14c304888ad7cd68bc4cd1fabd10931a719560a42a
+oid sha256:4fcd95de792c72a38d4c1e76a4e714a2c69ffc25f03172075efb432e40ec29cd
 size 8374456
diff --git a/cpp/tensorrt_llm/batch_manager/x86_64-linux-gnu/version.txt b/cpp/tensorrt_llm/batch_manager/x86_64-linux-gnu/version.txt
@@ -1,2 +1,2 @@
 e383212a40dca932c7b77bf4544dab80 libtensorrt_llm_ucx_wrapper.so
-d2efc6043262c896e262e8d8b97055af0f1f8b47 commit
+edf502396e4443f284a5fae6044402478cf457c1 commit
diff --git a/cpp/tensorrt_llm/executor/aarch64-linux-gnu/libtensorrt_llm_executor_static.a b/cpp/tensorrt_llm/executor/aarch64-linux-gnu/libtensorrt_llm_executor_static.a
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e2fbf80d02c115b9eeb2c18d24e3cb55f0c5404eba563591abeab7d223518df6
+oid sha256:6fd8ea50100bbbdc9d1d52d4b7e9a82f01583884eeb4d2703d537b6785c63ea7
 size 3102764
diff --git a/cpp/tensorrt_llm/executor/aarch64-linux-gnu/libtensorrt_llm_executor_static.pre_cxx11.a b/cpp/tensorrt_llm/executor/aarch64-linux-gnu/libtensorrt_llm_executor_static.pre_cxx11.a
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:27f9acc83aa72979834cc8c216cf06e6ab4e9b10a7d1c9928bac3721fef037bd
+oid sha256:cd30000142d1256991fa27644d86dcd12a4c017eab9345a88ac705914aba8d11
 size 3145744
diff --git a/cpp/tensorrt_llm/executor/aarch64-linux-gnu/version.txt b/cpp/tensorrt_llm/executor/aarch64-linux-gnu/version.txt
@@ -1,3 +1,3 @@
-61ab1a6d4c62ee2a648f6daa5083c4de libtensorrt_llm_executor_static.a
-2f2bc67944c45ce0965704da43c9b1c4 libtensorrt_llm_executor_static.pre_cxx11.a
-d2efc6043262c896e262e8d8b97055af0f1f8b47 commit
+1146671822817c690387dc77d775b8c7 libtensorrt_llm_executor_static.a
+8f7cb0047a0c2690497a97911a60ed6d libtensorrt_llm_executor_static.pre_cxx11.a
+edf502396e4443f284a5fae6044402478cf457c1 commit
diff --git a/cpp/tensorrt_llm/executor/x86_64-linux-gnu/libtensorrt_llm_executor_static.a b/cpp/tensorrt_llm/executor/x86_64-linux-gnu/libtensorrt_llm_executor_static.a
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae603dd0c585a7ee601fb6816ac2cdde674d5c49b96c7dce88de2bc67ea727bc
+oid sha256:22951d2bb0e5da2a1eb20ae0eb74690ddd57e7f1dd9545762eed1e0f468dd4a5
 size 3457520
diff --git a/cpp/tensorrt_llm/executor/x86_64-linux-gnu/libtensorrt_llm_executor_static.pre_cxx11.a b/cpp/tensorrt_llm/executor/x86_64-linux-gnu/libtensorrt_llm_executor_static.pre_cxx11.a
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7adc486890442df336e145b0ccf982bc733f1a9cc8116f7ce56f1769cf7b1154
+oid sha256:0c13a28fc903da20aad74aeb1c3d04a3b1bf91421fdbe85a16b3552b3e7e431b
 size 3448406
diff --git a/cpp/tensorrt_llm/executor/x86_64-linux-gnu/version.txt b/cpp/tensorrt_llm/executor/x86_64-linux-gnu/version.txt
@@ -1,3 +1,3 @@
-e5da8cc2936606dfb49f4417d6961060 libtensorrt_llm_executor_static.a
-ad5dfb89c2d719d99d67346828e92e25 libtensorrt_llm_executor_static.pre_cxx11.a
-d2efc6043262c896e262e8d8b97055af0f1f8b47 commit
+34a5173ddebafd3f1621af2717a92f54 libtensorrt_llm_executor_static.a
+34eacc123dc995815fbd1e68ec98f78b libtensorrt_llm_executor_static.pre_cxx11.a
+edf502396e4443f284a5fae6044402478cf457c1 commit
diff --git a/cpp/tensorrt_llm/kernels/decoderMaskedMultiheadAttention/decoderXQAImplJIT/nvrtcWrapper/aarch64-linux-gnu/version.txt b/cpp/tensorrt_llm/kernels/decoderMaskedMultiheadAttention/decoderXQAImplJIT/nvrtcWrapper/aarch64-linux-gnu/version.txt
@@ -1,2 +1,2 @@
 f3143205203b038b9dca6dd32cf02f59 libtensorrt_llm_nvrtc_wrapper.so
-d2efc6043262c896e262e8d8b97055af0f1f8b47 commit
+edf502396e4443f284a5fae6044402478cf457c1 commit
diff --git a/cpp/tensorrt_llm/kernels/decoderMaskedMultiheadAttention/decoderXQAImplJIT/nvrtcWrapper/x86_64-linux-gnu/version.txt b/cpp/tensorrt_llm/kernels/decoderMaskedMultiheadAttention/decoderXQAImplJIT/nvrtcWrapper/x86_64-linux-gnu/version.txt
@@ -1,2 +1,2 @@
 770ca93818f3f04837a67353e3f71fbc libtensorrt_llm_nvrtc_wrapper.so
-d2efc6043262c896e262e8d8b97055af0f1f8b47 commit
+edf502396e4443f284a5fae6044402478cf457c1 commit
diff --git a/cpp/tensorrt_llm/kernels/internal_cutlass_kernels/aarch64-linux-gnu/version.txt b/cpp/tensorrt_llm/kernels/internal_cutlass_kernels/aarch64-linux-gnu/version.txt
@@ -1,3 +1,3 @@
 6bf0ba4e9b8b1152a21316243d30bec6 libtensorrt_llm_internal_cutlass_kernels_static.a
 96f8a359c84a78ba415f4d98ef1c4e1d libtensorrt_llm_internal_cutlass_kernels_static.pre_cxx11.a
-d2efc6043262c896e262e8d8b97055af0f1f8b47 commit
+edf502396e4443f284a5fae6044402478cf457c1 commit
diff --git a/cpp/tensorrt_llm/kernels/internal_cutlass_kernels/x86_64-linux-gnu/libtensorrt_llm_internal_cutlass_kernels_static.a b/cpp/tensorrt_llm/kernels/internal_cutlass_kernels/x86_64-linux-gnu/libtensorrt_llm_internal_cutlass_kernels_static.a
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0b0f621e74dd506e49acd027dc09e9d2a3a6e0117ca0af68254841c02fb9c1dd
-size 68126454
+oid sha256:447838fe5c798098410a2cfed027aa38df847da2f725b9b8ccec57e73a1e194a
+size 68114502
diff --git a/cpp/tensorrt_llm/kernels/internal_cutlass_kernels/x86_64-linux-gnu/libtensorrt_llm_internal_cutlass_kernels_static.pre_cxx11.a b/cpp/tensorrt_llm/kernels/internal_cutlass_kernels/x86_64-linux-gnu/libtensorrt_llm_internal_cutlass_kernels_static.pre_cxx11.a
@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8448e105a5002148083bacc6c5066e017719ccd532c021f4c821df74fa0b763f
-size 68295728
+oid sha256:42b88e56cee5b9b81a66836add80ba79819afa24cbfd72140f4d62a244e3f960
+size 68295696
diff --git a/cpp/tensorrt_llm/kernels/internal_cutlass_kernels/x86_64-linux-gnu/version.txt b/cpp/tensorrt_llm/kernels/internal_cutlass_kernels/x86_64-linux-gnu/version.txt
@@ -1,3 +1,3 @@
-0b3322f5047dd4ee549211c2d15483c4 libtensorrt_llm_internal_cutlass_kernels_static.a
-502d4901fad6e648b8858051017c4cf2 libtensorrt_llm_internal_cutlass_kernels_static.pre_cxx11.a
-d2efc6043262c896e262e8d8b97055af0f1f8b47 commit
+4de75ffa1ff225422ba27f367175448f libtensorrt_llm_internal_cutlass_kernels_static.a
+e91d6c762f26c0b158eba8f376914e6e libtensorrt_llm_internal_cutlass_kernels_static.pre_cxx11.a
+edf502396e4443f284a5fae6044402478cf457c1 commit
diff --git a/docs/source/release-notes.md b/docs/source/release-notes.md
@@ -5,6 +5,12 @@
 All published functionality in the Release Notes has been fully tested and verified with known limitations documented. To share feedback about this release, access our [NVIDIA Developer Forum](https://forums.developer.nvidia.com/).
 
 
+## TensorRT-LLM Release 0.18.2
+
+### Key Features and Enhancements
+  - This update addresses known security issues. For the latest NVIDIA Vulnerability Disclosure Information visit https://www.nvidia.com/en-us/security/.
+
+
 ## TensorRT-LLM Release 0.18.1
 
 ### Key Features and Enhancements
diff --git a/examples/baichuan/requirements.txt b/examples/baichuan/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.15.0
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/bloom/requirements.txt b/examples/bloom/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/chatglm/requirements.txt b/examples/chatglm/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 evaluate~=0.4.1
 protobuf
diff --git a/examples/commandr/requirements.txt b/examples/commandr/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets==2.14.6
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/dbrx/requirements.txt b/examples/dbrx/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/deepseek_v1/requirements.txt b/examples/deepseek_v1/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.6
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/draft_target_model/requirements.txt b/examples/draft_target_model/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 rouge_score~=0.1.2
 sentencepiece>=0.1.99
diff --git a/examples/eagle/requirements.txt b/examples/eagle/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 rouge_score~=0.1.2
 SentencePiece~=0.1.99
diff --git a/examples/falcon/requirements.txt b/examples/falcon/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 transformers>=4.31.0
 datasets~=2.14.5
 evaluate~=0.4.1
diff --git a/examples/gemma/requirements.txt b/examples/gemma/requirements.txt
@@ -2,7 +2,7 @@
 # WAR the new posting of "nvidia-cudnn-cu12~=9.0".
 # "jax[cuda12_pip]~=0.4.19" specifies "nvidia-cudnn-cu12>=8.9" but actually requires "nvidia-cudnn-cu12~=8.9".
 nvidia-cudnn-cu12~=8.9; platform_machine == "x86_64"
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 flax~=0.8.0
 # jax[cuda12_pip]~=0.4.19; platform_system != "Windows"
 jax~=0.4.19; platform_system == "Windows"
diff --git a/examples/gpt/requirements.txt b/examples/gpt/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/gptj/requirements.txt b/examples/gptj/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/gptneox/requirements.txt b/examples/gptneox/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 rouge_score~=0.1.2
 evaluate~=0.4.1
diff --git a/examples/grok/requirements.txt b/examples/grok/requirements.txt
@@ -1,5 +1,5 @@
 -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets==2.14.6
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/internlm/requirements.txt b/examples/internlm/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets==2.14.5
 rouge_score~=0.1.2
 sentencepiece>=0.1.99
diff --git a/examples/jais/requirements.txt b/examples/jais/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/llama/requirements.txt b/examples/llama/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 transformers>=4.43.0
 datasets==2.14.6
 evaluate~=0.4.1
diff --git a/examples/lookahead/requirements.txt b/examples/lookahead/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 rouge_score~=0.1.2
 sentencepiece>=0.1.99
diff --git a/examples/mamba/requirements.txt b/examples/mamba/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 transformers>=4.39.0
 datasets~=2.14.5
 evaluate
diff --git a/examples/medusa/requirements.txt b/examples/medusa/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 rouge_score~=0.1.2
 sentencepiece>=0.1.99
diff --git a/examples/mixtral/requirements.txt b/examples/mixtral/requirements.txt
@@ -1,3 +1,3 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 transformers==4.38.2
 accelerate==0.25.0
diff --git a/examples/mpt/requirements.txt b/examples/mpt/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/nemotron/requirements.txt b/examples/nemotron/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 nemo-toolkit[all]==2.0.0rc1
 megatron-core==0.8.0
 datasets~=2.14.5
diff --git a/examples/opt/requirements.txt b/examples/opt/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/phi/requirements.txt b/examples/phi/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/prompt_lookup/requirements.txt b/examples/prompt_lookup/requirements.txt
@@ -1,5 +1,5 @@
 --extra-index-url https://pypi.nvidia.com
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 rouge_score~=0.1.2
 sentencepiece~=0.1.99
diff --git a/examples/quantization/requirements.txt b/examples/quantization/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets>=2.14.4
 nemo-toolkit[all]==2.0.0rc1
 rouge_score~=0.1.2
diff --git a/examples/qwen/requirements.txt b/examples/qwen/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.16.0
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/qwenvl/requirements.txt b/examples/qwenvl/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.16.0
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/recurrentgemma/requirements.txt b/examples/recurrentgemma/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 git+https://github.com/google-deepmind/recurrentgemma.git@8a32e365
 flax>=0.8.2
 jax~=0.4.23
diff --git a/examples/redrafter/requirements.txt b/examples/redrafter/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.14.5
 rouge_score~=0.1.2
 sentencepiece>=0.1.99
diff --git a/examples/skywork/requirements.txt b/examples/skywork/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets~=2.16.1
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/smaug/requirements.txt b/examples/smaug/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 datasets==2.14.6
 evaluate~=0.4.1
 rouge_score~=0.1.2
diff --git a/examples/whisper/requirements.txt b/examples/whisper/requirements.txt
@@ -1,4 +1,4 @@
-tensorrt_llm==0.18.1
+tensorrt_llm==0.18.2
 tiktoken
 datasets
 kaldialign
diff --git a/tensorrt_llm/executor.py b/tensorrt_llm/executor.py
diff --git a/tensorrt_llm/version.py b/tensorrt_llm/version.py

Original file line number	Diff line number	Diff line change
`@@ -1,2 +1,2 @@`
`1`	`1`	`9f9942768fd5b0cf5ed19860ad539dc9 libtensorrt_llm_ucx_wrapper.so`
`2`		`-d2efc6043262c896e262e8d8b97055af0f1f8b47 commit`
	`2`	`+edf502396e4443f284a5fae6044402478cf457c1 commit`
Original file line number	Diff line number	Diff line change
`@@ -1,2 +1,2 @@`
`1`	`1`	`e383212a40dca932c7b77bf4544dab80 libtensorrt_llm_ucx_wrapper.so`
`2`		`-d2efc6043262c896e262e8d8b97055af0f1f8b47 commit`
	`2`	`+edf502396e4443f284a5fae6044402478cf457c1 commit`