GitHub
Menu

Models / MiMo (Xiaomi)

MiMo-V2.5-Pro
serving recipe.

MiMo-V2.5-Pro is a 1T total, 42B active mixture of experts model with a 1M tokens context window. vLLM publishes a tested serving recipe for it, and Wingsmith reads that recipe when you register the weights.

Size
1T total, 42B activeMixture of experts
Context
1M tokensAs the recipe states it.
Memory floor
1224 GBSmallest published, whole replica.
vLLM
0.21.0 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
defaultFP81224 GBXiaomiMiMo/MiMo-V2.5-Pro

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
  • 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

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 XiaomiMiMo/MiMo-V2.5-Pro \
  --trust-remote-code \
  '--max-model-len auto' \
  --generation-config vllm \
  --tensor-parallel-size 8 \
  --gpu-memory-utilization 0.95 \
  --max-model-len auto \
  --tool-call-parser mimo \
  --enable-auto-tool-choice \
  --reasoning-parser mimo
Single-node baseline. Multi-node layouts add flags per strategy and role.

Features

  • Tool callingOn by defaultMiMo tool-call parser
  • ReasoningOn by defaultMiMo reasoning parser
  • Speculative decodingOpt-inMulti-Token Prediction

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:XiaomiMiMo/MiMo-V2.5-Pro"
    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 MiMo-V2.5-Pro need?The vLLM recipe publishes a minimum of 1224 GB across the whole replica for its FP8 variant. That is vLLM's own measurement, not a per-GPU figure.
  • Which parallelism layouts does the MiMo-V2.5-Pro recipe support?Tensor Parallel, Tensor + Expert Parallel and Prefill/Decode Disaggregation.
  • Can Wingsmith serve MiMo-V2.5-Pro?Wingsmith serves models through NVIDIA Dynamo on clusters with NVIDIA GPUs. Register XiaomiMiMo/MiMo-V2.5-Pro 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 XiaomiMiMo/MiMo-V2.5-Pro ↗ (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.