chore: bump trtllm to 1.3.0rc7 for Kimi K2.5 fix (#7408)

benchmarks/pyproject.toml

@@ -48,7 +48,7 @@ dependencies = [
     "pandas",
     "pydantic>=2",
     "tabulate",
-    # Satisfies vLLM 0.11.0 (>=4.55.2), vLLM 0.11.2 (>=4.56.0,<5), TRT-LLM 1.3.0rc5 (==4.57.1), SGLang 0.5.8 (==4.57.1)
+    # Satisfies vLLM 0.11.0 (>=4.55.2), vLLM 0.11.2 (>=4.56.0,<5), TRT-LLM 1.3.0rc7 (==4.57.1), SGLang 0.5.8 (==4.57.1)
     "transformers>=4.56.0",
 ]
 

container/context.yaml

@@ -96,10 +96,10 @@ trtllm:
   python_version: "3.12"
   index_url: https://pypi.nvidia.com/
   pip_wheel_dir: /tmp/trtllm_wheel/
-  pip_wheel: tensorrt-llm==1.3.0rc5.post1
+  pip_wheel: tensorrt-llm==1.3.0rc7
   trtllm_wheel_image: nvcr.io/nvidia/tensorrt-llm/release:${TENSORRTLLM_PIP_WHEEL#*==}
 
-  github_trtllm_commit: v1.3.0rc5.post1
+  github_trtllm_commit: v1.3.0rc7
   torch_version: 2.10.0a0+b4e4ee81d3.nv25.12
   torch_tensorrt_version: 2.10.0a0
   torchvision_version: 0.25.0a0+ca221243

container/deps/requirements.common.txt

@@ -22,7 +22,7 @@ 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 1.3.0rc5: ==4.57.1
+# - TensorRT-LLM 1.3.0rc7: ==4.57.1
 # - SGLang 0.5.8: ==4.57.1
 # Using >=4.56.0 to satisfy all frameworks
 transformers>=4.56.0

docs/reference/support-matrix.md

@@ -27,7 +27,7 @@ The following table shows the backend framework versions included with each Dyna
 
 | **Dynamo** | **SGLang** | **TensorRT-LLM** | **vLLM** | **NIXL** |
 | :--- | :--- | :--- | :--- | :--- |
-| **main (ToT)** | `0.5.9` | `1.3.0rc5.post1` | `0.17.1` | `0.10.1` |
+| **main (ToT)** | `0.5.9` | `1.3.0rc7` | `0.17.1` | `0.10.1` |
 | **v1.0.0** | `0.5.9` | `1.3.0rc5.post1` | `0.16.0` | `0.10.1` |
 | **v0.9.1** | `0.5.8` | `1.3.0rc3` | `0.14.1` | `0.9.0` |
 | **v0.9.0** | `0.5.8` | `1.3.0rc1` | `0.14.1` | `0.9.0` |

pyproject.toml

@@ -44,7 +44,7 @@ Repository = "https://github.com/ai-dynamo/dynamo.git"
 [project.optional-dependencies]
 trtllm =[
     "uvloop",
-    "tensorrt-llm==1.3.0rc5.post1",
+    "tensorrt-llm==1.3.0rc7",
 ]
 
 vllm = [