Evidence checked Nov 10, 2026

Verified Nov 10, 2026 via MiMo-V2-Flash: mimo-v2-flash-projectAI models

MiMo-V2-Flash

MiMo-V2-Flash is a Mixture-of-Experts language model from the XiaomiMiMo project with 309B total and 15B active parameters, designed for high-speed reasoning and agentic workflows according to its README.

Verified facts
8
Offers
1
Sources
2
Pricing
Free offer

Overview

What MiMo-V2-Flash covers

MiMo-V2-Flash is a Mixture-of-Experts language model with 309B total and 15B active parameters, designed, the README says, for high-speed reasoning and agentic workflows. The README describes a hybrid attention design that interleaves sliding window and global attention in a 5:1 ratio with a 128-token window, and a multi-token prediction module.

It publishes MiMo-V2-Flash and MiMo-V2-Flash-Base, both listed with 309B total parameters, 15B active and a 256k context length, and says the 3-layer MTP weights are also open-sourced.

Read the full editorial overview

Licence and running it

The repository licence file is the Apache License, Version 2.0. The Hugging Face model card for the weights declares the MIT licence in its metadata, so confirm which terms govern the weights. The README ends with a Use Restrictions notice listing uses the reader agrees not to make, including any military purpose, malicious cyber activity and autonomous high-risk actions without human oversight. The notice says it is informational and does not amend the applicable model licence. VendorClue did not assess its legal effect, so read it as well as the licence file.

For inference the README says the model supports FP8 mixed precision, recommends SGLang and gives a launch example with a tensor-parallel size of 8 and a context length of 262144. It does not discuss pricing.

Maintenance and what is not established

The repository has no GitHub releases, so this capture follows the default branch. Its latest commit is dated 29 October 2026 and the repository is not archived.

Nothing was installed or run. The benchmark tables compare the model with others under the publisher's own settings and were not checked.

Sources

MiMo-V2-Flash: mimo-v2-flash-project · Evidence: documentation · Not checked

Plans & prices

Prices as listed by MiMo-V2-Flash, before any tax or fee the pricing page does not state.

Scroll the table sideways for every column.

PlanRate period not statedCoverage limitsAvailability
Plan: Self-hosted
FreeTax: Confirm at checkoutUpfront: not established
ConditionsPublisher model card (Hugging Face README at pinned revision 8b60b288) captured 9 November 2026; supplier description, not a deployment or independent test. The card declares the MIT licence and states no price: zero applies to the downloadable open weights run on your own hardware. The project’s GitHub repository, captured separately, carries a LICENSE file that GitHub identifies as Apache-2.0 and its README names no licence, so confirm which terms govern the weights. GPUs, hosting, serving software and any hosted API (the README links a Xiaomi MiMo API Platform whose prices were not captured) have separate prices and terms.
Unknown fees: Compute infrastructure, hosting and third-party servicesVerified Nov 10, 2026
The model card has a Use Restrictions notice: the user agrees not to use the model or derivatives for unlawful, fraudulent or malicious purposes, to infringe third-party rights, to exploit or harm individuals or groups, or for any military purpose or application.
ConditionsPublisher model card (Hugging Face README at pinned revision 8b60b288) captured 9 November 2026; supplier description, not a deployment or independent test. Applies to the downloadable weights (Self-hosted plan). The quoted list continues in the card with malicious cyber activities, unauthorised handling of personal data and autonomous high-risk actions without human oversight, and ends by saying the notice is informational and does not amend the model licence; read the full notice. The card declares the MIT licence while GitHub identifies the repository LICENSE file as Apache-2.0 and the project README names no licence, so confirm the governing licence. Hosted services such as the Xiaomi MiMo API Platform have separate terms that were not captured.
Verified Nov 10, 2026
The README says MiMo-V2-Flash supports FP8 mixed-precision inference and recommends SGLang for optimal performance.
ConditionsDated publisher project README (GitHub, commit 941e5547) captured 9 November 2026; supplier description, not a deployment or independent test. No minimum hardware is stated in that sentence; the README’s example launch is a multi-GPU configuration, so size the hardware separately. Nothing was deployed.
Verified Nov 10, 2026

MiMo-V2-Flash: mimo-v2-flash-modelcard-20261109b · MiMo-V2-Flash: mimo-v2-flash-project

Documented facts

Each fact comes from official documentation and shows when it was verified.

  • Who it is for

    MiMo-V2-Flash is a Mixture-of-Experts language model from the XiaomiMiMo project with 309B total and 15B active parameters, designed for high-speed reasoning and agentic workflows according to its README.

    Verified Nov 8, 2026 · MiMo-V2-Flash: mimo-v2-flash-project

  • Coverage limits

    The README states that the MiMo-V2-Flash context window supports lengths of up to 256k.

    Verified Nov 8, 2026 · MiMo-V2-Flash: mimo-v2-flash-project

  • Provider

    Xiaomi

    Dated publisher project README (GitHub, commit 941e5547) captured 9 November 2026; supplier description, not a deployment or independent test. The README’s recommended system prompt says the assistant is developed by Xiaomi; hosted-service terms were not captured.

    Verified Nov 10, 2026 · MiMo-V2-Flash: mimo-v2-flash-project

  • API model identifier

    XiaomiMiMo/MiMo-V2-Flash

    Dated publisher project README (GitHub, commit 941e5547) captured 9 November 2026; supplier description, not a deployment or independent test. Hugging Face repository identifier used in the README’s SGLang launch command, not the name of a hosted API model; nothing was deployed.

    Verified Nov 10, 2026 · MiMo-V2-Flash: mimo-v2-flash-project

  • Context window

    256k tokens in the context window

    Dated publisher project README (GitHub, commit 941e5547) captured 9 November 2026; supplier description, not a deployment or independent test. The README states the length without naming the unit; its launch example sets a context length of 262144. No long-context test was run.

    Verified Nov 10, 2026 · MiMo-V2-Flash: mimo-v2-flash-project

Applies to the facts above: Dated publisher project README captured on 7 November 2026; supplier description, not a deployment or independent test. Infrastructure, accounts, third-party services and managed offerings have separate costs and terms.

Not confirmed in our sources

We have not confirmed these terms for MiMo-V2-Flash in the linked documentation. That does not mean MiMo-V2-Flash lacks them; confirm them before you rely on them.

  • Modalities
  • Maximum output
  • Input token rate
  • Output token rate

Sources & methodology

Facts come from the linked documentation. Documentation checks are not hands-on tests, and readable sources can have incomplete coverage.

MiMo-V2-Flash: mimo-v2-flash-projectraw.githubusercontent.com
Retrieved Nov 10, 2026
Retrieved Nov 10, 2026

Discovered an undocumented discrepancy or pricing shift? Submit a correction request.

Submit correction

MiMo-V2-Flash alternatives

Other AI models tools to weigh. Different billing units do not establish a cost winner, so compare plans side by side.

MiMo-V2-Flash comparisons

Compare two products in the same category using their documented terms and sources.