MiMo-V2.5

Current board #25Arena #85Open weights

Xiaomi·xiaomi/mimo-v2.5

MiMo-V2.5 is a language model published by Xiaomi. 6 OpenRouter endpoints are published in the latest snapshot. It accepts text, audio, image, video input and returns text. The published context limit is 1.1M tokens. The catalogue declares support for reasoning, tool use, structured outputs.

ModelCap Index · Current board

74.7#25

Measured · 84% support · score interval 74.7–81.3

Market Gravity

72.5

Secondary market signal

Input

$0.14

per 1M tokens

Output

$0.28

per 1M tokens

Context

1.1M

tokens

Providers

endpoints

Gateway spend

—

latest · Vercel share

Providers

Uptime measured over the last 30 minutes

Provider	Input	Output	Context	Quant	Uptime
GMICloudfp8Cheapest	$0.112	$0.224	1.1M	fp8	93.86%
Xiaomifp8	$0.14	$0.28	1.0M	fp8	99.70%
Parasailfp8	$0.14	$0.28	1.0M	fp8	99.53%
Venicefp8	$0.14	$0.28	1M	fp8	100.0%
Novitafp8	$0.168	$0.336	1.0M	fp8	98.70%
DeepInfrabf16	$0.40	$2.00	262K	bf16	98.79%

Hugging Face Inference Providers

Exact repository identity

Provider economics and telemetry published by the Hugging Face Router. These are operational signals, not benchmark results.

Provider	Input	Output	Context	TTFT	Throughput	Features
deepinfra live	$0.40	$2.00	262K	621 ms	52.0 tok/s	tools · structured

Reported evaluation leads (6)

Excluded from score

Structured rows reported in this exact model repository at the pinned revision below. They are useful research leads, but remain quarantined until dataset revision, harness, configuration and identity pass ModelCap’s independent admission review.

Dataset / task	Metric	Value	Revision	Provenance
claw-eval/Claw-Eval general	—	62.1	Unpinned `not published`	Reported Quarantined
claw-eval/Claw-Eval multimodal	—	23.8	Unpinned `not published`	Reported Quarantined
claw-eval/Claw-Eval multi_turn	—	63.2	Unpinned `not published`	Reported Quarantined
harborframework/terminal-bench-2.0 terminalbench_2	—	65.8	Unpinned `not published`	Reported Quarantined
InternScience/ResearchClawBench overall	—	16.91	Unpinned `not published`	Reported Quarantined
ScaleAI/SWE-bench_Pro SWE_Bench_Pro	—	56.1	Unpinned `not published`	Reported Quarantined

ModelCap Index · Current board

Methodology

74.7

Measured · 2 public benchmark observations across 2 boards

Current board

#25

Public rank basis: Measured (measured)
Index support: 84.0% · strong
Index interval: 74.7–81.3
Rank posterior: Not available
Top 5 / Top 10 probability: Not available / Not available
Posterior as of: Snapshot timestamp unavailable
Identity binding: Exact catalogue product · xiaomi/xiaomi/mimo-v2.5/fp8 · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata XiaomiMiMo/MiMo-V2.5@63651580ca774f8504f676040460aed3e1244ac1 (not the evaluation revision)
Direct benchmark observations: 2
Breadth-adjusted BES point estimate: 78.0

Bayesian Evidence Score · measured evidence input only

BES normalizes admitted public benchmark evidence for measured Index rows. Its legacy score and observables prior do not define the public language rank.

Observed capability: 78.0
Evidence status: confirmed · 83% mass
General preference: 77.4
Coding: 80.6
Agents & tools: —
Reasoning: —
Evidence breadth: 92%
Evidence coverage: 84%

Market and catalogue signals · secondary, never the language Index rank

Usage (OpenRouter popularity): 75.7
Liquidity (providers × uptime): 78.9
Open reach (HF downloads): 52.2
Surface (context / tools / modalities): 100.0
Freshness: 66.5

· OpenRouter popularity #4
· 6 live providers
· HF 30d downloads 405348
· Catalogue: context 1050000, reasoning, tools, modalities Text/Audio/Image/Video
· First seen 2026-04-22

These adoption and deployment observations remain context only. They do not change this model's ModelCap Index score or rank.

BES evidence selected · measured input only

Arena Overall77.4
Arena Coding80.6

The bounded 0–100 capability score blends each source’s competitive placement with its published achievement, then combines capability families. Evidence breadth adds a modest uncertainty adjustment; price and popularity are not part of benchmark-led rank.

Deployment readiness

Metadata index

86.2

strong · 94% metadata coverage

Not capability

Artifact reproducibility: 85.0
Access & legal clarity: 100.0
Deployability: 99.7
Evaluation provenance: 53.8

Missing: artifactReproducibility.base-lineage, evaluationProvenance.evaluation-dataset-revisions.

A metadata completeness and deployability index—not a safety certification, quality grade, or production-readiness claim.

Market activity

Venue coverage

Market Gravity: 72.5
OpenRouter weekly popularity: #4 of 298

Market Gravity uses OpenRouter’s full-catalogue popularity order. Sparse Vercel top-ten observations appear only when published and do not affect the score.

Arena preference

Status: Ranked
Source: Arena
Official rank: #85 of 385
Rating: 1434.0
95% confidence interval: 1430–1438
Votes: 44K
Source snapshot: 6 Aug 2026, 13:00 UTC
Category: overall
Identity match confidence: 100%

Market Gravity breakdown

Usage55%: 75.7
Liquidity25%: 78.9
Open reach15%: 52.2
Freshness5%: 66.5

Specification

Context window: 1.1M tokens
Max output: 131K tokens
Inputs: Text, Audio, Image, Video
Outputs: Text
Tokenizer: Other
Cached input tokens: $0.0028 / 1M
First seen on OpenRouter: 22 Apr 2026

Capabilities

Supported: Reasoning
Supported: Tool use
Supported: Structured output
Supported: Response format
Not supported: Moderated

Weights & access

Open weights. Downloadable weights are published under an OSI or free-culture license in the current repository metadata. This describes weight availability, not whether the full training stack qualifies as Open Source AI.

Access: Open weights
Licence: mit
Pinned revision: 63651580ca77
Parameters: 310.8B
Architecture: MiMoV2ForCausalLM
Model type: mimo_v2
Hugging Face downloads (30d): 405K
Downloads (all time): 953K
Likes: 389
Repository: XiaomiMiMo/MiMo-V2.5
Repository updated: 9 Jul 2026, 04:23 UTC