Skip to content
ModelCap

MiniMax M2.5

All-versions archive #85Arena #150Restricted license

MiniMax·minimax/minimax-m2.5

MiniMax M2.5 is a language model published by MiniMax. 10 OpenRouter endpoints are published in the latest snapshot. It accepts text input and returns text. The published context limit is 205K tokens. The catalogue declares support for reasoning, tool use, structured outputs.

ModelCap Index · All-versions archive
62.4#85
Measured · 91% support · score interval 62.4–67.8
Market Gravity
46.3
Secondary market signal

Superseded by MiniMax M3. It no longer holds a current-board position; its all-versions archive rank is #85.

Input
$0.22
per 1M tokens
Output
$0.90
per 1M tokens
Context
205K
tokens
Providers
10
endpoints
Gateway spend
—
latest · Vercel share

Providers

Uptime measured over the last 30 minutes
ProviderInputOutputContextQuantUptime
Inceptronfp8Cheapest$0.22$0.90197Kfp8100.0%
DigitalOcean$0.225$0.9066K—100.0%
Venice$0.27$0.95198K——
StreamLake$0.27$1.08200K—100.0%
AtlasCloudfp8$0.295$1.20197Kfp8100.0%
Friendli$0.30$1.20197K—100.0%
Minimaxfp8$0.30$1.20205Kfp8100.0%
Novitafp8$0.30$1.20205Kfp8100.0%
SiliconFlowfp8$0.30$1.20197Kfp8100.0%
Minimaxhighspeed$0.60$2.40205Kfp8—

Hugging Face Inference Providers

Exact repository identity

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

ProviderInputOutputContextTTFTThroughputFeatures
novita
live
$0.30$1.20205K961 ms65.8 tok/stools
featherless-ai
live
——————

Reported evaluation leads (10)

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 / taskMetricValueRevisionProvenance
cais/hle
hle
—19.4
Unpinned
not published
Reported
Quarantined
collinear-ai/yc-bench
medium
—230,465
Unpinned
not published
Reported
Quarantined
Idavidrein/gpqa
diamond
—85.2
Unpinned
not published
Reported
Quarantined
internlm/WildClawBench
overall
—27.1
Unpinned
not published
Reported
Quarantined
internlm/WildClawBench
avg_time
—542
Unpinned
not published
Reported
Quarantined
internlm/WildClawBench
avg_cost
—9.6
Unpinned
not published
Reported
Quarantined
mercor/apex-agents
apex-agents
—6.2
Unpinned
not published
Reported
Quarantined
ScaleAI/SWE-bench_Pro
SWE_Bench_Pro
—55.4
Unpinned
not published
Reported
Quarantined
SWE-bench/SWE-bench_Verified
swe_bench_%_resolved
—75.8
Unpinned
not published
Reported
Quarantined
TIGER-Lab/MMLU-Pro
mmlu_pro
—80.1
Unpinned
not published
Reported
Quarantined

Other versions (5)

ModelCap Index · All-versions archive

Methodology
62.4
Measured · 4 public benchmark observations across 4 boards
All-versions archive
#85
Public rank basis
Measured (measured)
Index support
91.2% · strong
Index interval
62.4–67.8
Rank posterior
Not available
Top 5 / Top 10 probability
Not available / Not available
Posterior as of
Snapshot timestamp unavailable
Identity binding
Exact catalogue product · minimax/minimax/minimax-m2.5/fp8 · aggregates disclosed benchmark configurations · endpoint configuration not claimed · artifact metadata MiniMaxAI/MiniMax-M2.5@f710177d938eff80b684d42c5aa84b382612f21f (not the evaluation revision)
Direct benchmark observations
4
Breadth-adjusted BES point estimate
65.1
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
65.1
Evidence status
confirmed · 95% mass
General preference
63.5
Coding
81.0
Agents & tools
—
Reasoning
35.3
Evidence breadth
95%
Evidence coverage
91%
Market and catalogue signals · secondary, never the language Index rank
Usage (OpenRouter popularity)
21.6
Liquidity (providers × uptime)
97.6
Open reach (HF downloads)
—
Surface (context / tools / modalities)
73.5
Freshness
51.1
  • · OpenRouter popularity #87
  • · 9 live providers
  • · Catalogue: context 204800, reasoning, tools
  • · First seen 2026-02-12

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

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
90.0
strong · 94% metadata coverage
Not capability
Artifact reproducibility
85.0
Access & legal clarity
92.5
Deployability
98.4
Evaluation provenance
82.3

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
46.3
OpenRouter weekly popularity
#87 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
#150 of 385
Rating
1389.9
95% confidence interval
1386–1394
Votes
41K
Source snapshot
6 Aug 2026, 13:00 UTC
Category
overall
Identity match confidence
100%

Market Gravity breakdown

Usage55%
21.6

OpenRouter weekly popularity across the full model catalogue.

Liquidity25%
97.6

Independent providers versus the model's open or closed cohort, weighted by uptime.

Open reach15%
50.0

Hugging Face 30-day downloads for open models; neutral for closed models.

Freshness5%
51.1

Time since first public availability, on a six-month half-life.

Specification

Context window
205K tokens
Max output
197K tokens
Inputs
Text
Outputs
Text
Tokenizer
Other
Cached input tokens
$0.05 / 1M
First seen on OpenRouter
12 Feb 2026

Capabilities

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

Weights & access

Restricted license. Downloadable weights use a model-specific or use-restricted license. Review the linked license before use, redistribution, or commercial deployment.

Access
Restricted license
Licence
modified-mit
Pinned revision
f710177d938e
Parameters
228.7B
Architecture
MiniMaxM2ForCausalLM
Model type
minimax_m2
Hugging Face downloads (30d)
719K
Downloads (all time)
4.3M
Likes
2K
Repository
MiniMaxAI/MiniMax-M2.5
Repository updated
10 Mar 2026, 13:42 UTC