Singularity Logs

THE MODEL, IN DETAIL / MiniMaxAI

MiniMax-M2.5

minimax/minimax-m2.5Open weightsReasoning

MiniMax-M2.5 is a SOTA large language model designed for real-world productivity. Trained in a diverse range of complex real-world digital working environments, M2.5 builds upon the coding expertise of M2.1...

OpenRouter
Source-linked model intelligenceCatalog added 12 Feb 2026Specifications observed 16 Oct 2026

GPQA Diamond

85.2%

Source-reported

Context window

204.8K

32.8K max output tokens

Input price

$0.27

USD / million input tokens

Output price

$0.95

USD / million output tokens

Inference providers

8

9 provider routes observed

Measured output speed

—

No model-level measurement

01 / PERFORMANCE

A clearer view of capability.

Read the score, understand the context, and explore where this model sits among its peers.

Performance fingerprint

Individual published accuracy scores · not a composite index

At least three published accuracy scores are needed for this view.

Performance meets price

GPQA Diamond · same published source and version

Higher score · lower cost70$075$0.287580$0.57585$0.862590$1.15Qwen3.5-397B-A17B GPQA Diamond: 88.4 Input: $0.5/M tokensKimi-K2.5 GPQA Diamond: 87.6 Input: $0.5/M tokensQwen3.5-122B-A10B GPQA Diamond: 86.6 Input: $0.26/M tokensGLM-5 GPQA Diamond: 86 Input: $1/M tokensGLM-4.7 GPQA Diamond: 85.7 Input: $0.6/M tokensQwen3.5-27B GPQA Diamond: 85.5 Input: $0.26/M tokensMiniMax-M2.5 GPQA Diamond: 85.2 Input: $0.27/M tokensMiniMax-M2.5Kimi-K2-Thinking GPQA Diamond: 84.5 Input: $0.6/M tokensQwen3.5-35B-A3B GPQA Diamond: 84.2 Input: $0.15/M tokensStep-3.5-Flash GPQA Diamond: 83.5 Input: $0.1/M tokensDeepSeek-V3.2 GPQA Diamond: 82.4 Input: $0.259/M tokensQwen3.5-9B GPQA Diamond: 81.7 Input: $0.1/M tokensInput price · USD / million tokens
Qwenmoonshotaizai-orgMiniMaxAIstepfun-aideepseek-ai

In good company

GPQA Diamond · OpenEvals / Hugging Face

88.4
Qwen
87.6
moonshotai
86.6
Qwen
86
zai-org
85.7
zai-org
85.5
Qwen
85.2
This model
84.5
moonshotai
84.2
Qwen
83.8
Nanbeige

Rank #7 in this published cohort. 27 models in the source cohort. Showing selected nearby models. Evaluation settings can differ.

Evidence, by task.

Coverage is not a capability score.

2published results
1 evaluation source
Reasoning2 evaluations
CodingAwaiting evidence
MathematicsAwaiting evidence
VisionAwaiting evidence
AgenticAwaiting evidence
Long contextAwaiting evidence

Documented features, task evaluations, and preference rankings describe different things. Missing coverage does not imply a model cannot perform a task.

Explore our methodology

02 / BENCHMARK EXPLORER

The numbers behind the model.

Original results, exact versions, and a direct path to the evidence behind every score.

2 published results
020406080100GPQA Diamond: 85.2% · Published snapshot 2026-03-16 · OpenEvals / Hugging Face85.2%GPQA DiamondHigher is betterHumanity’s Last Exam: 19.4% · Published snapshot 2026-03-16 · OpenEvals / Hugging Face19.4%Humanity’s Last …Higher is better

2 percentage-based evaluations. Each benchmark measures a different task; compare results within the same evaluation and protocol.

Published benchmark results for MiniMax-M2.5
BenchmarkScoreVersion & metricSourcePublishedEvidence
Published snapshot 2026-03-16Accuracy · Higher is betterOpenEvals / Hugging FaceSource-reported15 Mar 2026
Published snapshot 2026-03-16Accuracy · Higher is betterOpenEvals / Hugging FaceSource-reported15 Mar 2026

Artificial Analysis indices

Published in the OpenRouter catalog · source-defined index values

Intelligence Index22.8

Catalog observation: 16 Oct 2026. Index versions and evaluation settings are not supplied in this feed. These are not Singularity composite scores.

Artificial Analysis

Design Arena

Source-reported Elo across individual design categories

3d1,170
models arena · Rank #60 · 57.6% win rate
Codecategories1,209
models arena · Rank #55 · 56.8% win rate
Dataviz1,173
models arena · Rank #68 · 51.2% win rate
Gamedev1,180
models arena · Rank #60 · 55.5% win rate
Svg1,147
models arena · Rank #45 · 54.5% win rate
Uicomponent1,172
models arena · Rank #67 · 53.4% win rate
Website1,224
models arena · Rank #53 · 57.5% win rate

Catalog observation: 16 Oct 2026. Category ratings have separate comparison pools.

03 / COST & VALUE

Know what your workload costs.

Explore token prices, see where the spend goes, and model a workload of your own.

Input$0.27per million tokens
Output$0.95per million tokens
Cache read$0.03per million tokens
Cache write—per million tokens

Workload cost calculator

Change the numbers. See the trade-offs.

Token estimate only. Excludes cache writes, tools, and other provider fees. Actual token usage and provider limits vary.

Your estimated spend

1,000 requests · 11M tokens

$3.65estimated total
Uncached input$2.7
Cached input$0
Output$0.95
Per request$0.0037
Without cache discount$3.65
Estimated cache savings$0

04 / INFERENCE NETWORK

One model. Different ways to run it.

Compare inference routes by price, context, quantization, and reported availability.

8 providers · 9 routes
Provider routes for MiniMax-M2.5, observed 16 Oct 2026
Provider / routeInput / MOutput / MCache / MContextOutput limitSpeedLatencyAvailability · 1dRates
VVeniceQuantization not disclosed$0.27$0.95$0.03198K32.8K——97.45%
SStreamLakeQuantization not disclosed$0.27$1.08$0.027200K128K——100%
AAtlasCloudFP8$0.295$1.2$0.06196.6K176.9K——100%
GGMICloudFP8$0.3$1.2$0.03196.6K176.9K——100%
DDigitalOceanQuantization not disclosed$0.3$1.2$0.0665.5K59K——100%
FFriendliQuantization not disclosed$0.3$1.2$0.06196.6K176.9K——100%
NNovitaFP8$0.3$1.2$0.03204.8K131.1K——100%
MMinimaxFP8$0.3$1.2$0.03204.8K131.1K——100%
MMinimaxFP8 · priority$0.6$2.4$0.06204.8K131.1K——99%

Snapshot observed 16 Oct 2026. Speed and latency are the reported 30-minute median when available. This is a dated snapshot, not live monitoring.

Provider price comparison

USD per million tokens · input and output rates

InputOutput
Venice
Input
$0.27
Output
$0.95
StreamLake
Input
$0.27
Output
$1.08
AtlasCloudfp8
Input
$0.295
Output
$1.2
GMICloudfp8
Input
$0.3
Output
$1.2
DigitalOcean
Input
$0.3
Output
$1.2
Friendli
Input
$0.3
Output
$1.2
Novitafp8
Input
$0.3
Output
$1.2
Minimaxfp8
Input
$0.3
Output
$1.2
Minimaxfp8 · priority
Input
$0.6
Output
$2.4

Reported availability

Provider availability over the selected observation window

Venice97.45%
StreamLake100%
AtlasCloud · FP8100%
GMICloud · FP8100%
DigitalOcean100%
Friendli100%
Novita · FP8100%
Minimax · FP8100%
Minimax · priority · FP899%

Output speed

Reported 30-minute p50 · tokens per second

No provider throughput measurements were returned in this snapshot.

Response latency

Reported 30-minute p50 · seconds

No provider latency measurements were returned in this snapshot.

Context by provider

Actual route limits can differ from the catalog maximum

Venice198K
StreamLake200K
AtlasCloud · FP8196.6K
GMICloud · FP8196.6K
DigitalOcean65.5K
Friendli196.6K
Novita · FP8204.8K
Minimax · FP8204.8K
Minimax · priority · FP8204.8K

05 / CAPABILITIES & MODALITIES

What goes in. What comes out.

A map of documented interfaces and API features, separate from benchmark performance.

INPUT MODALITIES

Text
MiniMax-M2.5Reasoning supported

OUTPUT MODALITIES

Text

Feature support

Listed catalog API parameters and documented pricing features

ReasoningSupported
Tool callingSupported
Tool choiceSupported
Structured outputsSupported
Response formatSupported
Reasoning detailsSupported
Seed controlSupported
Stop sequencesSupported
Prompt cachingSupported
TemperatureSupported

Room to work

Context capacity and completion limits

204.8Ktotal context tokens
Context window
32.8Kmax output

Output tokens share the total context budget. The maximum completion length is a separate limit, and individual provider routes may have smaller limits.

Catalog context
204,800 tokens
Maximum output
32,768 tokens
Top-provider context
198,000 tokens

Features vary by route

Check the provider before depending on a specific API feature.

API feature support by provider
Provider routeReasoningToolsStructured outputResponse formatImplicit caching
VeniceSupportedSupportedNot listedNot listedSupported
StreamLakeSupportedSupportedSupportedSupportedSupported
AtlasCloud · FP8SupportedSupportedSupportedSupportedSupported
GMICloud · FP8SupportedSupportedNot listedSupportedSupported
DigitalOceanSupportedSupportedSupportedSupportedSupported
FriendliSupportedSupportedSupportedSupportedSupported
Novita · FP8SupportedSupportedNot listedSupportedSupported
Minimax · FP8SupportedSupportedNot listedSupportedSupported
Minimax · priority · FP8SupportedSupportedNot listedSupportedSupported

06 / UNDER THE HOOD

The complete specification sheet.

Exact identities, model architecture, limits, and the defaults that shape a request.

Identity & architecture

Developer
MiniMaxAI
Exact identity
MiniMaxAI/MiniMax-M2.5
Subject
Model
Access
Open weights
License
Not disclosed
Parameters
228.7 billion
Tokenizer
Other
Instruction format
Not disclosed

Reasoning & controls

Reasoning
Supported
Mandatory reasoning
Yes
Reasoning by default
Not documented
Published configuration
Not disclosed
Supported reasoning efforts
Not documented
Default reasoning effort
Not disclosed
API parameters
19
Provider moderation
No
Quantization variants
fp8

Dates & limits

Catalog added
12 Feb 2026
Knowledge cutoff
Not disclosed
Catalog expiration
Not documented
Specifications observed
16 Oct 2026
Providers observed
16 Oct 2026
Request limits
Provider-specific / not supplied
Canonical catalog ID
minimax/minimax-m2.5-20260211

API parameters & defaults

Availability is catalog-wide. Defaults can vary across providers.

frequency_penaltySupported · default not supplied
include_reasoningSupported · default not supplied
logit_biasSupported · default not supplied
logprobsSupported · default not supplied
max_tokensSupported · default not supplied
min_pSupported · default not supplied
presence_penaltySupported · default not supplied
reasoningSupported · default not supplied
repetition_penaltySupported · default not supplied
response_formatSupported · default not supplied
seedSupported · default not supplied
stopSupported · default not supplied
structured_outputsSupported · default not supplied
temperatureDefault: 1
tool_choiceSupported · default not supplied
toolsSupported · default not supplied
top_kSupported · default not supplied
top_logprobsSupported · default not supplied
top_pDefault: 0.95

07 / BUILD WITH THIS MODEL

From exploration to your first request.

Use the documented catalog identifier and keep your API key in an environment variable.

Make an API request

const response = await fetch(
  "https://openrouter.ai/api/v1/chat/completions",
  {
    method: "POST",
    headers: {
      Authorization: `Bearer ${process.env.OPENROUTER_API_KEY}`,
      "Content-Type": "application/json",
    },
    body: JSON.stringify({
      "model": "minimax/minimax-m2.5",
      "messages": [
        {
          "role": "user",
          "content": "Explain the trade-offs in this design."
        }
      ]
    }),
  },
);
if (!response.ok) throw new Error(`HTTP ${response.status}`);
const data = await response.json();
console.log(data.choices[0].message.content);

08 / PROVENANCE & HISTORY

Every number has a story.

Follow the source, check the observation date, and distinguish published results from model specifications.

The observation trail

Recorded events, in chronological order

Added to the model catalog

Catalog creation date, as recorded by OpenRouter.

GPQA Diamond result published

85.2% · OpenEvals / Hugging Face

Humanity’s Last Exam result published

19.4% · OpenEvals / Hugging Face

Model specifications observed

Catalog pricing, limits, and supported parameters captured.

Provider routes observed

9 inference routes in this snapshot.

Observation dates describe when data was captured. They do not represent newly run evaluations or continuous performance monitoring.

Published observation ledger

Historical evidence, with source and version preserved.

Published observation history for MiniMax-M2.5
ObservedBenchmarkScoreVersionSourceVerification
13 Oct 2026Humanity’s Last ExamPublished snapshot 2026-03-16OpenEvals / Hugging FaceSource-reported
13 Oct 2026GPQA DiamondPublished snapshot 2026-03-16OpenEvals / Hugging FaceSource-reported

09 / A FEW MORE DETAILS

Questions worth asking.

Quick answers, drawn from this model’s connected specifications.

How much does MiniMax-M2.5 cost?

Catalog starting rates are $0.27 per million input tokens and $0.95 per million output tokens. Cache-read pricing is $0.03 per million tokens. Compare provider routes and service tiers above for their specific prices.

How much context can it handle?

204,800 tokens in the catalog context window. The catalog lists a maximum of 32,768 completion tokens. Provider limits can differ.

Does it support tools and structured output?

Tool calling: listed as supported. Structured outputs: listed as supported. See the provider feature matrix for route-specific differences.

How should I interpret these benchmark results?

Read the benchmark name, exact version, metric, source, and model configuration together. Different evaluation settings can change the result. Preference ratings and source-defined composite indices remain separate from task accuracy.

Are these prices and provider measurements live?

This page uses dated source snapshots. Catalog data was observed on 16 Oct 2026. Provider routes were observed on 16 Oct 2026. Missing measurements are not estimated or treated as zero.

KEEP EXPLORING

More from MiniMaxAI.

Explore related identities from the same developer, or take them into a side-by-side comparison.

THE RIGHT MODEL STARTS WITH THE RIGHT CONTEXT.

Put the details side by side.

Explore the trade-offs in a direct model comparison.

Compare MiniMax-M2.5
Source-backed evidence. Different tasks. A clearer picture.
AboutMethodologyBenchmark dashboard