Skip to content

MiniMax · Chat

MiniMax M3.1 Flash

MiniMax 🇨🇳ChatOne of its sources says the weights aren't publicly available.
From Empiriolabs's listing:
Fast multimodal coding and agent model with 1M context, adjustable reasoning depth, image and video input, function calling, and built-in web search.
llminfo IQ
134.5
Max context
1M
Listings
111

Evidence and sources

International API storefront (minimax.io) is Singapore-based; principal operating entity 上海稀宇科技有限公司 (Shanghai).

🤗 MiniMaxAI/MiniMax-M3.1-Flash Model architecture, as its sources state it llminfo IQ 134.5, ranked among chat models. From 60 benchmarks. Plenty of data behind it.How llminfo IQ is computedCoding IQ 118.7, 95% interval 109–128. From 6 benchmark families across 3 sources. A fair amount of data behind it. Uses the model's best published results. It has its own scale, separate from llminfo IQ.How Coding IQ is computedGood value (60.2). Cheaper per task than its benchmark results predict. From 17 benchmarks with measured task costs. Plenty of data behind it.How Value for money worksReleased 19 Oct 2026Maker HQ: CN (source)
All rankings: #7 on OpenRouter this week #31 on OpenRouter this month #41 on OpenRouter today #184 on NanoGPT recentlySources: LMArena data, EmpirioLabs data, Infercom data, ModelRunner.ai data, flex.ai API data, llminfo researchPrimary data from LMArena from EmpirioLabs: name, primary mode, description from Infercom: architecture from ModelRunner.ai: weights availability from flex.ai API: hugging face id from llminfo research: release date
Tags from its sources:textvisioninstructtoolsreasoningopen-sourcemoelong-contextcodingagentic

Advantages and trade-offs

Compared with 10 chat models, picked mainly by popularity and IQ

Advantages

Trade-offs

No open weights

The weights aren't public, so you can't run it yourself.

We only make a claim when there's enough data on models of the same type. Updated 22 Nov, 06:07 UTC.Prices come from the model's maker where they sell it, otherwise their official partner, otherwise the market median. This comparison uses 9 maker's prices.

Input & output

as stated by the model's providers
Text
Images
Video
Text

Capabilities

Supported operations and provider-stated features

VisionReasoningFunction callingStructured outputPrompt cachingWeb search

Listings

 MiniMaxMiniMax-M3.1-FlashSources: LiteLLM data, MiniMax API Pricing, Models.dev data1M$0.15cache read $0.015$0.60
 Opencode Gominimax-m3.1-flash-free · freeSource: Models.dev data1M$0.00$0.00
 TenstorrentMiniMaxAI/MiniMax-M3.1-FlashSource: Models.dev data1M$0.05cache read $0.005$0.20
 Eden AIInference by MiniMaxminimax/MiniMax-M3.1-FlashSource: Models.dev data1M$0.075cache read $0.0075$0.30
 NanoGPTInference by RelaceRegions: UNKminimax/minimax-m3.1-flash · fp4Source: NanoGPTNot published by the provider$0.0945cache read $0.0105$0.525
 LlmgatewayInference by Inference Netminimax-m3.1-flashSource: LLMGateway dataNot published by the provider$0.10cache read $0.005$0.50
 Empiriolabsminimax-m3-1-flashSources: EmpirioLabs data, LiteLLM data1M$0.1125cache read $0.0113$0.45
 GmicloudMiniMaxAI/MiniMax-M3.1-Flash · fp8Sources: GMI Cloud data, LiteLLM data1M$0.1125cache read $0.0113$0.45
 Hugging FaceInference by WaferMiniMaxAI/MiniMax-M3.1-FlashSource: Hugging Face Inference Providers data1M$0.135$0.54
 Kilominimax/minimax-m3.1-flashSources: Kilo AI Gateway data, Models.dev data1M$0.135cache read $0.0135$0.54

13 listings without a price are hidden.

Included in plans

MiniMax M Plan GoMiniMaxSubscription$22/moPlan page
MiniMax M Plan ExploreMiniMaxSubscription$55/moPlan page
MiniMax M Plan BuildMiniMaxSubscription$132/moPlan page

Benchmarks

Named effort levels run from lowest to highest. Select a result to see its source details.

Agentic

Agents' Last Exam · Overall ScoreMiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · official ALE Overall full-set leaderboard result · CLI-only runs excluded · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Arena AgentSource: LMArena data · Nov 2026
Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · OpenCode 1.17.18 · MiniMax M2.7 judge
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Finance Agent v2.1Vals AI October 20 model evaluation · official MiniMax API · temperature 1 / top_p 0.95 / compute effort max · output cap inconsistent across the page · Reported by Vals AI · via BenchBoard docs
OSWorld 2.02 results
Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · OSWorld 2.0 0808 Full release · highest reported at release · metric subtype unstated
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
PostTrainBench v1.1MiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · Claude Code 2.1.251 · 512K context · otherwise official setup · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
Source: OpenRouter benchmark exports
19 Oct 2026 version · 90-day results
Across 17 providers lowest 66% median 70% highest 74.7%OpenRouter's automatic routing 66.7%
Show results from 17 providers and OpenRouter's automatic routingHide results from 17 providers and OpenRouter's automatic routing
  • Relace
  • Nebius
  • Inferact vLLM
  • Together
  • Fireworks
  • ModelRun
  • Novita
  • SambaNova
  • StreamLake
  • Parasail
  • InferenceNet
  • Minimax
  • Wafer
  • Tenstorrent
  • DigitalOcean
  • GMICloud
  • OpenRouter's automatic routing
  • Makora
Tax Agent BenchVals AI October 20 model evaluation · official MiniMax API · temperature 1 / top_p 0.95 / compute effort max · output cap inconsistent across the page · Reported by Vals AI · via BenchBoard docs
Reported by Vals AI · via BenchBoard docs
Vals AI October 20 model evaluation · official MiniMax API · temperature 1 / top_p 0.95 / compute effort max · output cap inconsistent across the page
Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · Codex 0.149.1 · provider-reported result, distinct from maintainer leaderboard
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
ValsSource: Vals AI

Science

GPQA Diamond18 results
Source: OpenRouter benchmark exports
19 Oct 2026 version · 90-day results
Across 17 providers lowest 84.9% median 87.3% highest 89.3%OpenRouter's automatic routing 87.6%
Show results from 17 providers and OpenRouter's automatic routingHide results from 17 providers and OpenRouter's automatic routing
  • Novita
  • Inferact vLLM
  • StreamLake
  • Minimax
  • GMICloud
  • OpenRouter's automatic routing
  • Makora
  • Together
  • Nebius
  • DigitalOcean
  • ModelRun
  • SambaNova
  • Relace
  • InferenceNet
  • Wafer
  • Tenstorrent
  • Parasail
  • Fireworks

Coding

CyberGym2 results
Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · Claude Code 2.1.263 · 1M context · launch prose rounds to 82%
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
NL2RepoSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · Claude Code 2.1.220 · 200K context · otherwise official setup
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
Vibe Code Bench v1.1Vals AI October 20 model evaluation · official MiniMax API · temperature 1 / top_p 0.95 / compute effort max · output cap inconsistent across the page · Reported by Vals AI · via BenchBoard docs

Reasoning

Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · v1.0.6 · 600 publicly released tasks
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
BabyVision2 results
Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · with tools · Claude Code 2.1.285 · minimum image side 1,500 px
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated
GDP.pdfSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Humanity's Last ExamMiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · without tools · text-only subset / 2,158 questions · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.

Vision

CharXiv · RQMiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · reasoning questions · GPT-5.5 judge · code-interpreter setting not stated · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
CharXiv-RSource: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
MiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · official implementation · 256K
Source: LLM Stats data · Reported by the model's developer. The source relaying it did not verify it. claim
Setting not stated

Human preference

Math

Other

DeepSWE v1.1MiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · mini-SWE-agent 2.4.5 · 512K context · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
EMB v1.1Vals AI October 20 model evaluation · official MiniMax API · temperature 1 / top_p 0.95 / compute effort max · output cap inconsistent across the page · Reported by Vals AI · via BenchBoard docs
GDP.pdf · Artificial AnalysisMiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · Artificial Analysis GDP.pdf AA result reproduced in provider card · evaluation harness not restated · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
HLAB-Vals-1.1Vals AI October 20 model evaluation · official MiniMax API · temperature 1 / top_p 0.95 / compute effort max · output cap inconsistent across the page · Reported by Vals AI · via BenchBoard docs
NL2Repo-BenchMiniMax M3.1 Flash official-weights model-card run · reasoning effort, sampling, and run count not stated for this benchmark · Claude Code 2.1.233 · 512K context · reward hacking = 0 · two inconsistent tasks corrected · Reported by MiniMax · via BenchBoard docs · Reported by the model's developer. The source relaying it did not verify it.
SuperCLUESource: SuperCLUE data
Vals10 related benchmarks · 10 results
Finance Index · October 2026
Reported by Vals AI · via BenchBoard docs
Vals AI October 20 model evaluation · official MiniMax API · temperature 1 / top_p 0.95 / compute effort max · output cap inconsistent across the page
Show 4 more resultsHide 4 results
Index v2.2
Reported by Vals AI · via BenchBoard docs
Vals AI October 20 model evaluation · official MiniMax API · temperature 1 / top_p 0.95 / compute effort max · output cap inconsistent across the page

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links. · Hugging Face official benchmark leaderboards; licence varies by benchmark · LMArena, CC BY 4.0 (licence) · Data by LLM Stats · Baromètre de l’écriture IA, Johannes Eva; public numeric ranking data, © 2026 Johannes Eva, no results licence stated · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · Clinical Benchmarks (clinicalbenchmarks.ai), CC BY 4.0; fractions converted to percentages; source documents retain their own terms (licence)

Also known as: accounts/fireworks/models/minimax-m3p1-flash, cline-pass/minimax-m3.1-flash, empiriolabs/minimax-m3-1-flash, fireworks/minimax-m3.1-flash, fireworks_ai/accounts/fireworks/models/minimax-m3p1-flash, flexai/minimax-m3.1-flash, gmi/minimaxai/minimax-m3.1-flash, minimax m3.1 flash, minimax-m3-1-flash, minimax-m3.1-flash, minimax-m3.1-flash-free, minimax-m3.1-flash@eu, minimax/m3.1-flash, minimax/minimax-m3.1-flash, minimaxai/minimax-m3.1-flash, modelrunner-ai/minimax-m3.1-flash, nebius/minimax-m3.1-flash, nebius/minimaxai/minimax-m3.1-flash, novita/minimax-m3.1-flash, nvidia_nim/minimaxai/minimax-m3.1-flash, opencode_go/minimax-m3.1-flash, parasail-minimax-m31-flash, together-ai/minimax-m3.1-flash, together_ai/minimaxai/minimax-m3.1-flash

Like this one. Tap or hover any dotted value, label or icon to see where it comes from and how it is measured. open details when tapped or hovered. How this data works