Skip to main content

Model profile

MiniMax M2.7

MiniMaxSupersededReleased Mar 18, 2026
Data verified
Superseded:MiniMax has released newer models in this line —MiniMax M3
Overall Score
64.11Public #36 of 200Verified #29 of 99
Arena Elo
1418
Eligible category ranks
2of 8
Price (1M tokens)
$0.3 in / $1.2 out
API pricing
Speed
45tok/s
Context
200K

Evidence coverage

35 of 323 tracked benchmarks are published. 14 are verified and 21 provisional. 7 of 8 categories are measured.

Updated July 23, 2026Methodology
Published / tracked
35 / 323
Verified
14
Provisional
21
Categories with evidence
7 / 8

Evidence by category

  • Agentic11 benchmarks
    Mixed evidence
  • Coding11 benchmarks
    Mixed evidence
  • Reasoning2 benchmarks
    Reported
  • Knowledge8 benchmarks
    Reported
  • Math1 benchmark
    Reported
  • Multilingual0 benchmarks
    Not measured
  • Multimodal1 benchmark
    Reported
  • Inst. Following1 benchmark
    Reported
Open WeightSelf-hostNon-Reasoning
Confidence:
Low
base

MiniMax M2.7 ranks #36 out of 200 models on the public leaderboard with an overall score of 64.11/100. It also ranks #29 out of 99 on the verified leaderboard. While not a frontier model, it offers specific advantages depending on the use case.

MiniMax M2.7 is a open weight model with a 200K token context window. It processes queries without explicit chain-of-thought reasoning, offering faster response times and lower token usage.

BenchLM links it directly to MiniMax M2.5 as the earlier related model in that lineage. This profile currently has 35 of 323 tracked benchmarks. BenchLM only exposes non-generated benchmark rows publicly, so missing categories stay blank until a sourced evaluation is available.

Its strongest category is Coding (#78), while its weakest is Agentic (#109). This performance profile makes it particularly well-suited for software development and code generation tasks.

Peer position

Exact provisional scores and ranks for the closest listed peers. A score can appear before a model clears the evidence threshold for a rank, so equal scores can have different rank states.

Range 63.564.75

  1. Gemini 3.5 Flash
    Google
    #3364.75
    Gemini 3.5 Flash is #33 with a score of 64.75.
    Compare
  2. Claude Opus 4.5
    Anthropic
    #3464.22
    Claude Opus 4.5 is #34 with a score of 64.22.
    Compare
  3. Claude Opus 4.6 (Adaptive)
    Anthropic
    #3564.18
    Claude Opus 4.6 (Adaptive) is #35 with a score of 64.18.
    Compare
  4. MiniMax M2.7Current model
    MiniMax
    #3664.11
    MiniMax M2.7 is #36 with a score of 64.11.
  5. GLM-5.2
    Z.AI
    #3763.96
    GLM-5.2 is #37 with a score of 63.96.
    Compare
  6. GPT-5.5 Pro
    OpenAI
    #3863.69
    GPT-5.5 Pro is #38 with a score of 63.69.
    Compare
  7. GLM-5V-Turbo
    Z.AI
    #3963.5
    GLM-5V-Turbo is #39 with a score of 63.5.
    Compare

Category percentile

More

Relative position among models eligible for each sourced category. A higher percentile means a stronger position within that category's ranked cohort; 100 is highest.

  1. Coding36%
    Eligible cohort rank #78 of 122Category score 47.6
  2. Agentic8%
    Eligible cohort rank #109 of 119Category score 34.8

Category evidence

Scores and ranks appear only where this model has published benchmark evidence. Categories without displayable source records remain not measured.

Category scores, ranks, weighting, benchmark coverage, and evidence status
CategoryScore
AgenticRank #109 of 119Percentile 8thWeight 22%11 benchmarksMixed sources34.8
CodingRank #78 of 122Percentile 36thWeight 20%11 benchmarksMixed sources47.6
ReasoningWeight 17%2 benchmarksReportedScore pending
KnowledgeRank Not rankedWeight 12%8 benchmarksReported28.0
MathRank Not rankedWeight 5%1 benchmarkReported81.3
MultilingualWeight 7%0 benchmarksNot measuredNot measured
MultimodalWeight 12%1 benchmarkReportedScore pending
Inst. FollowingRank Not rankedWeight 5%1 benchmarkReported75.7

Chatbot Arena performance

Scroll horizontally to inspect confidence intervals and vote counts.

Chatbot Arena Elo, confidence interval, and vote count by evaluation view
ViewEloConfidence intervalVotes
Text Overall1418±4.249,135
Coding1480±6.314,249
Math1424±12.12,601
Instruction Following1410±6.015,705
Creative Writing1365±8.07,485
Multi-turn1428±7.58,762
Hard Prompts1444±5.031,808
Hard Prompts (English)1460±6.215,078
Longer Query1435±5.919,813

Benchmark Details

Rows below have a displayable published verification record. Each source link and provenance note remains in the page HTML while its category is closed. Source-unverified manual rows and generated rows stay hidden.

Agentic11 benchmarks
Terminal-Bench 2.0Provider exact
57%Weighted 38%
Source: MiniMax M2.7Provenance: Provider exact
τ²-bench resultsReported

τ²-Bench Tool-Agent-User Evaluation

84.8%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
ToolathlonProvider exact
46.3%Display only
Source: MiniMax M2.7Provenance: Provider exact
MLE-Bench LiteProvider exact
66.6%Display only
Source: MiniMax M2.7Provenance: Provider exact
MM-ClawBenchProvider exact
62.7%Display only
Source: MiniMax M2.7Provenance: Provider exact
Claw-EvalBenchmark exact
48.7%Display only
Source: Claw-Eval leaderboardProvenance: Claw-Eval reports this model as minimax_m27 in the official 2026-05-09 leaderboard snapshot. BenchLM stores the primary Pass^3 value on the local Claw-Eval display key.
AA Agentic IndexReported

Artificial Analysis Agentic Index

25.6%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
APEX-Agents-AAReported
10.6%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
GDPval-AAReported

GDPval-AA normalized

32.9%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
GDPval-AAReported
1158Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Gert LabsBenchmark exact

Gert Labs Composite Game Benchmark

40.40%Display only
Source: Gert Labs rankingsProvenance: Gert Labs reports this composite leaderboard score in the public rankings API. BenchLM scales the source gscore from 0-1 to 0-100 and stores it as a display-only agentic benchmark.
Coding11 benchmarks
SWE-RebenchBenchmark exact
51.9%Weighted 20%
Source: SWE-Rebench leaderboardProvenance: Live default SWE-Rebench leaderboard lists MiniMax M2.7 at 51.9% resolved rate.
SWE-bench ProProvider exact
56.2%Weighted 10%
Source: MiniMax M2.7Provenance: Provider exact
SWE-bench Verified*Secondary exact

SWE-bench Verified (mini-swe-agent-v2)

75.4%Display only
Source: Arcee Trinity-Large-Thinking comparison tableProvenance: Secondary exact
SWE MultilingualProvider exact
76.5%Display only
Source: MiniMax M2.7Provenance: Provider exact
Multi-SWE BenchProvider exact
52.7%Display only
Source: MiniMax M2.7Provenance: Provider exact
VIBE-ProProvider exact
55.6%Display only
Source: MiniMax M2.7Provenance: Provider exact
NL2RepoProvider exact
39.8%Display only
Source: MiniMax M2.7Provenance: Provider exact
Vibe Code BenchBenchmark exact

Vibe Code Bench v1.1

27.04%Display only
Source: Vals AI: Vibe Code Bench v1.1Provenance: Vals Vibe Code Bench v1.1 reports this exact row under minimax/MiniMax-M2.7; BenchLM stores it on the local vibeCodeBench key.
React Native EvalsBenchmark exact
71.4%Display only
Source: React Native Evals leaderboardProvenance: React Native Evals reports this exact overall score for Minimax M2.7 in the public dashboard run finished on 2026-04-28.
AA Coding IndexReported

Artificial Analysis Coding Index

52.6%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-SciCodeReported

Artificial Analysis SciCode

47.0%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Reasoning2 benchmarks
AA-LCRReported

Artificial Analysis Long Context Reasoning

68.7%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
CritPtReported

Critical Physics Tasks

0.6%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Knowledge8 benchmarks
GPQA-DSecondary exact

GPQA Diamond

87.0%Display only
Source: Arcee Trinity-Large-Thinking comparison tableProvenance: Secondary exact
MMLU-Pro (Arcee)Secondary exact

MMLU-Pro first-party comparison snapshot

80.8%Display only
Source: Arcee Trinity-Large-Thinking comparison tableProvenance: Secondary exact
Artificial Analysis Intelligence IndexReported
38.1%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-GPQA DiamondReported

Artificial Analysis GPQA Diamond

87.4%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-HLEReported

Artificial Analysis Humanity's Last Exam

28.1%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-Omniscience IndexReported

Artificial Analysis Omniscience Index

0.7%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-Omniscience AccuracyReported

Artificial Analysis Omniscience Accuracy

26.1%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
AA-Omniscience Hallucination RateReported

Artificial Analysis Omniscience Hallucination Rate

34.4%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.
Math1 benchmark
AIME25 (Arcee)Secondary exact

AIME25 first-party comparison snapshot

80.0%Display only
Source: Arcee Trinity-Large-Thinking comparison tableProvenance: Secondary exact
Multimodal1 benchmark
Design Arena WebsiteReported

Design Arena Website Elo

1275Display only
Source: OpenRouter model benchmarksProvenance: Display-only Design Arena Website Elo synced from OpenRouter model benchmark metadata. It is excluded from BenchLM weighted scoring.
Inst. Following1 benchmark
AA-IFBenchReported

Artificial Analysis IFBench

75.7%Display only
Source: Artificial Analysis model benchmarksProvenance: Display-only row synced from the current Artificial Analysis model payload. It is excluded from BenchLM weighted scoring.

MiniMax M2.7 Family

Base entry

Related Earlier Model

MiniMax M2.5

Frequently Asked Questions

How does MiniMax M2.7 perform overall in AI benchmarks?

MiniMax M2.7 has 35 published benchmark scores on BenchLM, but it does not yet have enough non-generated coverage to receive a global overall rank.

Is MiniMax M2.7 good for knowledge and understanding?

MiniMax M2.7 has visible benchmark coverage in knowledge and understanding, but BenchLM does not currently assign it a global category rank there.

Is MiniMax M2.7 good for coding and programming?

MiniMax M2.7 ranks #78 out of 122 models in coding and programming benchmarks with an average score of 47.6. There are stronger options in this category.

Is MiniMax M2.7 good for mathematics?

MiniMax M2.7 has visible benchmark coverage in mathematics, but BenchLM does not currently assign it a global category rank there.

Is MiniMax M2.7 good for reasoning and logic?

MiniMax M2.7 has visible benchmark coverage in reasoning and logic, but BenchLM does not currently assign it a global category rank there.

Is MiniMax M2.7 good for agentic tool use and computer tasks?

MiniMax M2.7 ranks #109 out of 119 models in agentic tool use and computer tasks benchmarks with an average score of 34.8. There are stronger options in this category.

Is MiniMax M2.7 good for multimodal and grounded tasks?

MiniMax M2.7 has visible benchmark coverage in multimodal and grounded tasks, but BenchLM does not currently assign it a global category rank there.

Is MiniMax M2.7 good for instruction following?

MiniMax M2.7 has visible benchmark coverage in instruction following, but BenchLM does not currently assign it a global category rank there.

Is MiniMax M2.7 open source?

Yes, MiniMax M2.7 is an open weight model created by MiniMax, meaning it can be downloaded and run locally or fine-tuned for specific use cases.

Does MiniMax M2.7 have full benchmark coverage on BenchLM?

Not yet. MiniMax M2.7 currently has 35 published benchmark scores out of the 323 benchmarks BenchLM tracks. BenchLM only exposes non-generated public benchmark rows, so missing categories stay blank until a sourced evaluation is available.

What is the context window size of MiniMax M2.7?

MiniMax M2.7 has a published context window of 200K, which determines how much text it can process in a single interaction.

Last updated: July 23, 2026 · Runtime metrics stay blank until BenchLM has a sourced snapshot.

Choose with this week’s evidence

Join 2,000+ readers for ranking moves, new releases, pricing changes, and the evidence behind them.

Free. One email per week.