chore: update dynamo and nixl versions for 0.3.0 (#1240)

nv-anants · nv-anants · commit 50971b7d079a · 2025-05-29T10:34:09.000-07:00
diff --git a/Cargo.lock b/Cargo.lock
diff --git a/Cargo.toml b/Cargo.toml
@@ -28,7 +28,7 @@ members = [
 resolver = "3"
 
 [workspace.package]
-version = "0.2.1"
+version = "0.3.0"
 edition = "2021"
 description = "Dynamo Inference Framework"
 authors = ["NVIDIA Inc. <sw-dl-dynamo@nvidia.com>"]
@@ -39,9 +39,9 @@ keywords = ["llm", "genai", "inference", "nvidia", "distributed", "dynamo"]
 
 [workspace.dependencies]
 # Local crates
-dynamo-runtime = { path = "lib/runtime", version = "0.2.1" }
-dynamo-llm = { path = "lib/llm", version = "0.2.1" }
-dynamo-tokens = { path = "lib/tokens", version = "0.2.1" }
+dynamo-runtime = { path = "lib/runtime", version = "0.3.0" }
+dynamo-llm = { path = "lib/llm", version = "0.3.0" }
+dynamo-tokens = { path = "lib/tokens", version = "0.3.0" }
 
 # External dependencies
 anyhow = { version = "1" }
diff --git a/container/Dockerfile.vllm b/container/Dockerfile.vllm
@@ -169,7 +169,7 @@ RUN uv pip install /workspace/wheels/nixl/*.whl
 ARG VLLM_REF="0.8.4"
 ARG VLLM_PATCH="vllm_v${VLLM_REF}-dynamo-kv-disagg-patch.patch"
 ARG VLLM_PATCHED_PACKAGE_NAME="ai_dynamo_vllm"
-ARG VLLM_PATCHED_PACKAGE_VERSION="0.8.4.post1"
+ARG VLLM_PATCHED_PACKAGE_VERSION="0.8.4.post2"
 ARG VLLM_MAX_JOBS=4
 RUN --mount=type=bind,source=./container/deps/,target=/tmp/deps \
     --mount=type=cache,target=/root/.cache/uv \
diff --git a/container/build.sh b/container/build.sh
@@ -109,7 +109,7 @@ NONE_BASE_IMAGE_TAG="24.04"
 SGLANG_BASE_IMAGE="nvcr.io/nvidia/cuda-dl-base"
 SGLANG_BASE_IMAGE_TAG="25.01-cuda12.8-devel-ubuntu24.04"
 
-NIXL_COMMIT=78695c2900cd7fff506764377386592dfc98e87e
+NIXL_COMMIT=f531404be4866d85ed618b3baf4008c636798d63
 NIXL_REPO=ai-dynamo/nixl.git
 
 NO_CACHE=""
diff --git a/docs/support_matrix.md b/docs/support_matrix.md
@@ -62,21 +62,21 @@ For **Linux**, the **ARM64** support is experimental and may have limitations. W
 ### Runtime Dependency
 | **Python Package** | **Version**   | glibc version        | CUDA Version |
 |--------------------|---------------|----------------------|--------------|
-| ai-dynamo          |    0.2.1      |     >=2.28           |              |
-| ai-dynamo-runtime  |    0.2.1      |     >=2.28           |              |
-| ai-dynamo-vllm     |  0.8.4.post1* | >=2.28 (recommended) |                    |
-| NIXL               |    0.2.1      |     >=2.27           | >=11.8      |
+| ai-dynamo          |    0.3.0      |     >=2.28           |              |
+| ai-dynamo-runtime  |    0.3.0      |     >=2.28           |              |
+| ai-dynamo-vllm     |  0.8.4.post2* | >=2.28 (recommended) |                    |
+| NIXL               |    0.3.0      |     >=2.27           | >=11.8      |
 
 ### Build Dependency
 | **Build Dependency** | **Version** |
 |----------------------|-------------|
 | **Base Container**   |    [25.03](https://catalog.ngc.nvidia.com/orgs/nvidia/containers/cuda-dl-base/tags)    |
-| **ai-dynamo-vllm**   |0.8.4.post1* |
+| **ai-dynamo-vllm**   |0.8.4.post2* |
 | **TensorRT-LLM**     |    0.19.0** |
-| **NIXL**             |    0.2.1    |
+| **NIXL**             |    0.3.0    |
 
 > **Note**:
-> - *ai-dynamo-vllm v0.8.4.post1 is a customized patch of v0.8.4 from vLLM.
+> - *ai-dynamo-vllm v0.8.4.post2 is a customized patch of v0.8.4 from vLLM.
 > - **Specific versions of TensorRT-LLM supported by Dynamo are subject to change.
 
 
diff --git a/lib/bindings/python/Cargo.lock b/lib/bindings/python/Cargo.lock
diff --git a/lib/bindings/python/Cargo.toml b/lib/bindings/python/Cargo.toml
@@ -19,7 +19,7 @@
 
 [package]
 name = "dynamo-py3"
-version = "0.2.1"
+version = "0.3.0"
 edition = "2021"
 authors = ["NVIDIA"]
 license = "Apache-2.0"
@@ -75,3 +75,4 @@ pyo3-async-runtimes = { version = "0.23.0", default-features = false, features =
 pythonize = "0.23"
 
 dlpark = { version = "0.5", features = ["pyo3", "half"], optional = true }
+
diff --git a/lib/llm/Cargo.toml b/lib/llm/Cargo.toml
@@ -81,7 +81,7 @@ regex = "1"
 rayon = "1"
 
 # block_manager
-nixl-sys = { version = "0.2.1-rc.3", optional = true }
+nixl-sys = { version = "0.3.0-rc.2", optional = true }
 cudarc = { version = "0.16.2", features = ["cuda-12020"], optional = true }
 ndarray = { version = "0.16", optional = true }
 nix = { version = "0.26", optional = true }
diff --git a/lib/runtime/examples/Cargo.lock b/lib/runtime/examples/Cargo.lock
diff --git a/lib/runtime/examples/Cargo.toml b/lib/runtime/examples/Cargo.toml
@@ -21,7 +21,7 @@ members = [
 resolver = "3"
 
 [workspace.package]
-version = "0.2.1"
+version = "0.3.0"
 edition = "2021"
 authors = ["NVIDIA"]
 license = "Apache-2.0"
diff --git a/pyproject.toml b/pyproject.toml
@@ -15,7 +15,7 @@
 
 [project]
 name = "ai-dynamo"
-version = "0.2.1"
+version = "0.3.0"
 description = "Distributed Inference Framework"
 readme = "README.md"
 authors = [
@@ -29,7 +29,7 @@ dependencies = [
     "bentoml==1.4.8",
     "types-psutil==7.0.0.20250218",
     "kubernetes==32.0.1",
-    "ai-dynamo-runtime==0.2.1",
+    "ai-dynamo-runtime==0.3.0",
     "fastapi==0.115.6",
     "distro",
     "typer",
