XiaomiMiMo V2.6 Pro

External Link

MiMo-V2.6-Pro is Xiaomi's open-weight, native omni-modal sparse mixture-of-experts model for reasoning and agentic workloads.

Token context
1,049K tokens
Inputs
Text, Image, Video, Audio
Outputs
Text
Released
Sep 21, 2026

Providers & Pricing

Estimate workload cost →
ProviderInput / 1MOutput / 1MCache read / 1MCache write / 1MObservedEvidence
$0.435$0.87$0.0036Sep 23, 2026Provider-reported
$0.435$0.87$0.0036Sep 22, 2026Provider-reported

At a Glance

Model Facts

Identity

Status
active
Developer ID
mimo-v2.6-pro
Released
Sep 21, 2026
Version
MiMo-V2.6-Pro
License
MIT

Capacity

Token context
1,049K tokens
Maximum output
131K tokens
Total parameters
1T
Active parameters
42B
Knowledge cutoff
Not reported

Interface and Access

Inputs
Text, Image, Video, Audio
Outputs
Text
API available
Yes
Open weights
Yes
Self-hostable
Yes
Reasoning
Yes
Vision
Yes
Tool calling
Yes
Capabilities
chat, generation, reasoning, tools

This page represents one developer model product. Dated API snapshots, serving endpoints, pricing, regions, quantizations, and service tiers attach as versioned aliases or provider details and never create duplicate public model records.

Version Lineage

PredecessorUnknown
SuccessorsUnknown
Aliasesmimo-v2.6-pro, MiMo-V2.6-Pro-RL, XiaomiMiMo/MiMo-V2.6-Pro-RL, xiaomi/mimo-v2.6-pro-20260921

Explicit Unknowns

  • Tokenizer
  • Knowledge cutoff

Primary Evidence

Sources and Observation Date

Comparable Peers

Related Model Comparisons

All comparisons →
APairBContext
vscross-developer peersimage, text
vscross-developer peersimage, text
vscross-developer peerstext
vscross-developer peersimage, text
vscross-developer peersimage, text
vscross-developer peersimage, text, video
vscross-developer peersimage, text
vscross-developer peersimage, text
vscross-developer peerstext
vscross-developer peerstext
vscross-developer peersimage, text, video
vscross-developer peersimage, text
vscross-developer peersimage, text
vscross-developer peersimage, text, video
vscross-developer peersimage, text
vscross-developer peerstext
vscross-developer peerstext
vscross-developer peerstext
vscross-developer peerstext
vscross-developer peerstext
vscross-developer peerstext
vscross-developer peersimage, text

Questions

MiMo V2.6 Pro FAQs

What is MiMo V2.6 Pro?+

MiMo-V2.6-Pro is Xiaomi's open-weight, native omni-modal sparse mixture-of-experts model for reasoning and agentic workloads. It is developed by Xiaomi and its current sourced lifecycle status is active.

What inputs and outputs does MiMo V2.6 Pro support?+

MiMo V2.6 Pro's current record lists Text, Image, Video, and Audio as input and Text as output.

How much context does MiMo V2.6 Pro support?+

The current record lists a 1,049K-token context window and a maximum output of 131K tokens.

Is MiMo V2.6 Pro available through an API?+

MiMo V2.6 Pro is marked as API-available. The directly sourced provider records currently include Openrouter and Xiaomi MiMo.

How large is MiMo V2.6 Pro?+

The current record lists 1T total parameters and 42B active parameters. Its architecture is Sparse mixture of experts.

What capabilities and tasks does MiMo V2.6 Pro support?+

MiMo V2.6 Pro's recorded capabilities are chat, generation, reasoning, and tools. Its supported tasks are audio, image, text, and video.

How much does MiMo V2.6 Pro cost through an API?+

The lowest directly sourced prices currently attached to MiMo V2.6 Pro are $0.435 per million input tokens through Openrouter and $0.87 per million output tokens through Openrouter. Prices are provider-specific and should be checked against each cited observation date.

Can MiMo V2.6 Pro be self-hosted?+

MiMo V2.6 Pro is an open-weight model. Self-hosting is supported, and the recorded license is MIT.

Does MiMo V2.6 Pro have related versions or aliases?+

No predecessor is recorded. No successor is recorded. Recorded aliases are mimo-v2.6-pro, MiMo-V2.6-Pro-RL, XiaomiMiMo/MiMo-V2.6-Pro-RL, and xiaomi/mimo-v2.6-pro-20260921.

Which models can MiMo V2.6 Pro be compared with?+

The current compatible peer set includes GPT-6 Astra, Claude Fable 5.1, Gemini 3.1 Pro, DeepSeek V4.1 Flash, and Grok 4.6. Compatibility requires shared sourced modalities and tasks; it does not imply that one model is better. Browse model comparisons

Send Feedback