GitHub
Menu

Models / Qwen

Qwen3-Coder-480B-A35B-Instruct
serving recipe.

Qwen3-Coder-480B-A35B-Instruct is a 480B total, 35B active mixture of experts model with a 256K tokens context window. vLLM publishes a tested serving recipe for it, and Wingsmith reads that recipe when you register the weights.

Size
480B total, 35B activeMixture of experts
Context
256K tokensAs the recipe states it.
Memory floor
288 GBSmallest published, whole replica.
vLLM
0.10.0 or laterMinimum version in the recipe.

01 / Memory

Precisions and memory floors.

Each variant is a precision vLLM tested, sometimes published as its own repository.

VariantPrecisionMemory floorWeights
defaultBF161152 GBQwen/Qwen3-Coder-480B-A35B-Instruct
fp8FP8576 GBQwen/Qwen3-Coder-480B-A35B-Instruct-FP8
nvfp4NVFP4288 GBnvidia/Qwen3-Coder-480B-A35B-Instruct-NVFP4

Floors are the minimum the vLLM recipe publishes across the whole replica, from vLLM's own runs. The control plane compares them with its own sizing from the model config.

02 / Parallelism

Layouts the recipe supports.

  • Tensor ParallelFits inside one node's interconnect. single_node_tp
  • Tensor + Expert ParallelFits inside one node's interconnect. single_node_tep
  • Data + Expert ParallelFits inside one node's interconnect. single_node_dep
  • Multi-Node Tensor ParallelSpans nodes. multi_node_tp
  • Multi-Node TP + Pipeline ParallelSpans nodes. multi_node_tp_pp
  • Multi-Node Data + Expert ParallelSpans nodes. multi_node_dep
  • Multi-Node Tensor + Expert ParallelSpans nodes. multi_node_tep
  • Prefill/Decode DisaggregationSpans nodes. pd_cluster

03 / Hardware

Where vLLM ran it.

Node shapes the recipe reports an end-to-end run on.

  • H200 · 8 GPUs
  • MI300X · 8 GPUs
  • MI325X · 8 GPUs
  • MI355X · 8 GPUs
  • TPU v7 (Ironwood) · 1 GPU

Wingsmith runs models through NVIDIA Dynamo, so it serves on NVIDIA GPUs. Other vendors are listed as the recipe states them.

04 / Flags

The tested baseline.

Base flags, then the default variant, then the features that are on by default, as the recipe composes them.

vllm serve Qwen/Qwen3-Coder-480B-A35B-Instruct \
  --enable-auto-tool-choice \
  --tool-call-parser qwen3_coder
Single-node baseline. Multi-node layouts add flags per strategy and role.

Features

  • Tool callingOn by defaultEnable automatic tool choice with Qwen3 Coder parser

Dependencies

  • Optional: opt into DeepGEMM FP8 MoE kernels for extra throughput (optional)

05 / Serve

Serve it with Wingsmith.

From a slug to an OpenAI-compatible endpoint, across your clusters.

  1. Register the weights

    In the console, or through the API. Resolving a slug proposes an artifact and writes nothing.

    curl "http://localhost:8000/v1/artifacts/resolve?source=hf:Qwen/Qwen3-Coder-480B-A35B-Instruct"
    Against the local control plane from the quickstart, where admin auth is off.

  2. Read the draft

    The draft carries this recipe as upstream evidence: the tested flags, the precision and memory floor, the parallelism layouts, and the hardware vLLM verified, with links back to the recipe. Nothing from it is applied without you, and the draft warns when the plane's own sizing and the published floor differ by a quarter or more.

  3. Create the workload

    Choose the engine, device shape and scale. The plane picks the clusters, and each cluster's operator renders the deployment as a Dynamo graph.

  4. Give it an endpoint

    One model name for clients, routed by the gateway to deployments on every cluster serving it.

Run the control plane locally →

Questions

Common questions.

  • How much GPU memory does Qwen3-Coder-480B-A35B-Instruct need?The vLLM recipe publishes a minimum of 288 GB across the whole replica for its NVFP4 variant; BF16 1152 GB, FP8 576 GB. That is vLLM's own measurement, not a per-GPU figure.
  • Which parallelism layouts does the Qwen3-Coder-480B-A35B-Instruct recipe support?Tensor Parallel, Tensor + Expert Parallel, Data + Expert Parallel, Multi-Node Tensor Parallel, Multi-Node TP + Pipeline Parallel, Multi-Node Data + Expert Parallel, Multi-Node Tensor + Expert Parallel and Prefill/Decode Disaggregation.
  • Can Wingsmith serve Qwen3-Coder-480B-A35B-Instruct?Wingsmith serves models through NVIDIA Dynamo on clusters with NVIDIA GPUs. Register Qwen/Qwen3-Coder-480B-A35B-Instruct and the plane attaches this vLLM recipe to the draft as evidence: its tested flags, precision and memory floor, for you to review rather than applied on your behalf.

From the vLLM recipe for Qwen/Qwen3-Coder-480B-A35B-Instruct ↗ (snapshot at 4bcc189), Apache 2.0.

Every model

Status

Open source,
under Apache 2.0.

Wingsmith is pre-release. The first public release, and everything after it, will be under Apache 2.0.

View on GitHubRead the docs Run it locally

Built and not built

Built

Multi-cluster placement and routing, Dynamo deployments, device inventory, endpoints and quotas, the console, the CLI, and three measurement tools.

Not built

Published images and charts, per-user roles and SSO, batch and training workloads, and a Helm chart for the control plane.

Pre-release. No tagged release, images, or charts yet, and APIs and protocols can change.