chore: TRTLLM 1.2.0rc6 (#5017)

Signed-off-by: Nate Mailhot <nmailhot@nvidia.com>

benchmarks/pyproject.toml

@@ -46,7 +46,7 @@ dependencies = [
     "pydantic>=2",
     "tabulate",
     "types-tabulate",
-    # Satisfies vLLM 0.11.0 (>=4.55.2), vLLM 0.11.2 (>=4.56.0,<5), TRT-LLM 1.2.0rc5 (==4.56.0), SGLang 0.5.6.post2 (==4.57.1)
+    # Satisfies vLLM 0.11.0 (>=4.55.2), vLLM 0.11.2 (>=4.56.0,<5), TRT-LLM 1.2.0rc6 (==4.56.0), SGLang 0.5.6.post2 (==4.57.1)
     "transformers>=4.56.0,<=4.57.1",
     "pytest-mypy",
 ]

container/Dockerfile.trtllm

@@ -543,7 +543,7 @@ RUN if [ "$HAS_TRTLLM_CONTEXT" = "1" ]; then \
         sed -i 's/pip3 install/uv pip install/g' /tmp/install_tensorrt.sh && \
         bash /tmp/install_tensorrt.sh && \
         # Install TensorRT-LLM wheel from the provided index URL, allow dependencies from PyPI
-        # TRTLLM 1.2.0rc5 has issues installing from pypi with uv, installing from direct wheel link works best
+        # TRTLLM 1.2.0rc6 has issues installing from pypi with uv, installing from direct wheel link works best
         # explicitly installing triton 3.5.0 as trtllm only lists triton as dependency on x64_64 for some reason
         if echo "${TENSORRTLLM_PIP_WHEEL}" | grep -q '^tensorrt-llm=='; then \
             TRTLLM_VERSION=$(echo "${TENSORRTLLM_PIP_WHEEL}" | sed -E 's/tensorrt-llm==([0-9a-zA-Z.+-]+).*/\1/'); \

container/build.sh

@@ -89,7 +89,7 @@ DEFAULT_TENSORRTLLM_PIP_WHEEL_DIR="/tmp/trtllm_wheel/"
 # TensorRT-LLM commit to use for building the trtllm wheel if not provided.
 # Important Note: This commit is not used in our CI pipeline. See the CI
 # variables to learn how to run a pipeline with a specific commit.
-DEFAULT_EXPERIMENTAL_TRTLLM_COMMIT="e4c707845ff58fcc0b1d87afb4dd0e64885c780a" # 1.2.0rc5
+DEFAULT_EXPERIMENTAL_TRTLLM_COMMIT="9ba14263db0045ed3fa0860f949b5ce320107eb3" # 1.2.0rc6
 TRTLLM_COMMIT=""
 TRTLLM_USE_NIXL_KVCACHE_EXPERIMENTAL="0"
 TRTLLM_GIT_URL=""
@@ -98,7 +98,7 @@ TRTLLM_GIT_URL=""
 DEFAULT_TENSORRTLLM_INDEX_URL="https://pypi.nvidia.com/"
 # TODO: Remove the version specification from here and use the ai-dynamo[trtllm] package.
 # Need to update the Dockerfile.trtllm to use the ai-dynamo[trtllm] package.
-DEFAULT_TENSORRTLLM_PIP_WHEEL="tensorrt-llm==1.2.0rc5"
+DEFAULT_TENSORRTLLM_PIP_WHEEL="tensorrt-llm==1.2.0rc6"
 TENSORRTLLM_PIP_WHEEL=""
 
 VLLM_BASE_IMAGE="nvcr.io/nvidia/cuda-dl-base"

container/deps/requirements.txt

@@ -52,7 +52,7 @@ tensorboard>=2.19.0,<2.21.0
 tensorboardX==2.6.2.2
 # Transformers version constraint for container builds
 # - vLLM 0.11.0: >=4.55.2, vLLM 0.11.2: >=4.56.0,<5
-# - TensorRT-LLM 1.2.0rc5: ==4.56.0
+# - TensorRT-LLM 1.2.0rc6: ==4.56.0
 # - SGLang 0.5.6.post2: ==4.57.1
 # Using >=4.56.0 and <=4.57.1 to satisfy all frameworks
 transformers>=4.56.0,<=4.57.1

docs/reference/support-matrix.md

@@ -63,7 +63,7 @@ The following table shows the dependency versions included with each Dynamo rele
 | **Dependency** | **main (ToT)** | **v0.8.0 (unreleased)** | **v0.7.1** | **v0.7.0.post1** | **v0.7.0** |
 | :------------- | :------------- | :---------------------- | :--------- | :--------------- | :--------- |
 | SGLang         | 0.5.6.post2    | 0.5.6.post2             | 0.5.3.post4| 0.5.3.post4      | 0.5.3.post4|
-| TensorRT-LLM   | 1.2.0rc5       | 1.2.0rc6                | 1.2.0rc3   | 1.2.0rc3         | 1.2.0rc2   |
+| TensorRT-LLM   | 1.2.0rc6       | 1.2.0rc6                | 1.2.0rc3   | 1.2.0rc3         | 1.2.0rc2   |
 | vLLM           | 0.12.0         | 0.12.0                  | 0.11.0     | 0.11.0           | 0.11.0     |
 | NIXL           | 0.8.0          | 0.8.0                   | 0.8.0      | 0.8.0            | 0.8.0      |
 

pyproject.toml

@@ -50,7 +50,7 @@ Repository = "https://github.com/ai-dynamo/dynamo.git"
 [project.optional-dependencies]
 trtllm =[
     "uvloop",
-    "tensorrt-llm==1.2.0rc5",
+    "tensorrt-llm==1.2.0rc6",
 ]
 
 vllm = [