VerdictPal
Join

Models / MiniMax

MiniMax M3

MiniMax · Released 2026-05-31

MiniMax's May 2026 open-weights release. Terminal-Bench 2.0 66.0% — the #10 score globally and the strongest non-frontier-tier result on the public ledger. 1M context, $0.40 / $2.00 per 1M tokens.

Default open-weights model for self-hosted agent infrastructure. Cheaper than DeepSeek V4 Pro Max on output ($2 vs $1.60 — well, slightly higher), with comparable Terminal-Bench scores.

#37 of 168 on AA Index · current snapshot
45 AA Index · 17 benchmark rows · 8 sourcesPosition is benchmark-specific — not a cross-family or cross-source ranking.
Composite63
Context1M
Input / 1M$0.40
Output / 1M$2.0
Knowledge cutoff2026-03-01
Output speed114 t/s
TTFT (default API)0.95s
Statussolid

Family profile

Best published score in each covered benchmark family.

79/100 avg
Knowledge
85
Reasoning
93
Coding
78
Agentic
66
Long context
80
Tool use
76
Human preference
75

7 tested benchmark families

Editor's note
MiniMax M3 is the dark horse of the open-weights tier — Terminal-Bench 2.0 66.0% puts it ahead of several proprietary models. The 1M context is the real differentiator for long-document work.
Benchmark placements

Where MiniMax M3 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.

BenchmarkFamilyRankScoreSourceDate
IFBenchReasoning#3/ 12783Artificial Analysis2026-09-01
AA-LCRLong context#5/ 17980Artificial Analysis2026-09-01
MCP AtlasTool use#10/ 1069MCP Atlas leaderboard2026-06-09
LMArena (Chatbot Arena)Human preference#12/ 1375LMArena leaderboard2026-05-25
GPQA DiamondReasoning#13/ 18593Artificial Analysis2026-09-01
BFCLTool use#14/ 2076Berkeley Function Calling Leaderboard V42026-04-15
SWE-bench VerifiedCoding#17/ 2378SWE-bench Verified leaderboard2026-06-09
LiveBenchReasoning#18/ 2160LiveBench leaderboard2026-04-15
LongBench v2Long context#18/ 1946LongBench v2 leaderboard2026-04-01
MMLUKnowledge#22/ 2285MiniMax M32026-05-31
AA output speedPerformance#32/ 187Artificial Analysis2026-09-01
Humanity's Last ExamReasoning#39/ 18039Artificial Analysis2026-09-01
Artificial Analysis Intelligence IndexReasoning#46/ 18745Artificial Analysis2026-09-01
Terminal-BenchAgentic#54/ 18466Terminal-Bench 2.0 leaderboard2026-05-20
AA time to first tokenPerformance#67/ 186Artificial Analysis2026-09-01
τ³-Bench BankingAgentic#69/ 10615Artificial Analysis2026-09-01
SciCodeCoding#80/ 18045Artificial Analysis2026-09-01
Family context

The three highest-scoring models with pages in each capability family. Where MiniMax M3 shows up, it's highlighted.

KnowledgeMiniMax M3 · 85
  1. 1
    GPT 5.5
    91
  2. 2
    Claude Opus 4.8
    91
  3. 3
    Claude Sonnet 4.6
    91
ReasoningMiniMax M3 · 93
  1. 1
    Claude Mythos Preview
    95
  2. 2
    Gemini 3.7 Flash
    95
  3. 3
    Grok 4.6
    95
CodingMiniMax M3 · 78
  1. 1
    GPT 5.5
    96
  2. 2
    Claude Opus 4.8
    96
  3. 3
    Claude Opus 4.6
    96
AgenticMiniMax M3 · 66
  1. 1
    GPT 5.5
    100
  2. 2
    Claude Opus 4.8
    100
  3. 3
    Claude Fable 5
    100
Long contextMiniMax M3 · 80
  1. 1
    Muse Spark 1.2
    83
  2. 2
    Kimi K3
    83
  3. 3
    Gemini 3.5 Flash
    81
Tool useMiniMax M3 · 76
  1. 1
    Claude Sonnet 4.6
    89
  2. 2
    GPT 5.5
    87
  3. 3
    Claude Opus 4.8
    87
Human preferenceMiniMax M3 · 75
  1. 1
    GPT 5.5
    95
  2. 2
    Claude Opus 4.8
    92
  3. 3
    Gemini 3.5 Flash
    91
Newest receipts

Benchmark rows added to the public ledger for MiniMax M3 in the last 120 days. Older rows live in the full table below.

14 recent
BenchmarkFamilyScoreSourceDays ago
AA-LCRLong context80.33%Artificial Analysis5
AA output speedPerformance114 t/sArtificial Analysis5
AA time to first tokenPerformance0.95sArtificial Analysis5
Artificial Analysis Intelligence IndexReasoning45.4%Artificial Analysis5
GPQA DiamondReasoning92.9%Artificial Analysis5
Humanity's Last ExamReasoning39%Artificial Analysis5
IFBenchReasoning82.86%Artificial Analysis5
SciCodeCoding45.4%Artificial Analysis5
τ³-Bench BankingAgentic15.26%Artificial Analysis5
SWE-bench VerifiedCoding78.4%SWE-bench Verified leaderboard89
MCP AtlasTool use68.9%MCP Atlas leaderboard89
MMLUKnowledge84.8%MiniMax M398
LMArena (Chatbot Arena)Human preference1376LMArena leaderboard104
Terminal-BenchAgentic66%Terminal-Bench 2.0 leaderboard109
Family coverage

A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.

Knowledge85
Reasoning93
Coding78
Agentic66
Long context80
Tool use76
Human preference75
Price vs. performance
0255075100$0.10$0.30$1$3$10$30input USD / 1M tokens · log scale →← intelligence

Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.

AnthropicOpenAIMistralAlibabaDeepSeekGoogleMoonshotMetaMiniMaxxAIZhipuXiaomiOtherMeituanCohereNVIDIA

Full benchmark ledger

Every catalog benchmark for MiniMax M3. Scores link to the original source; gaps mean no public row exists yet.

17 sourced rows

17 of 37 catalog benchmarks have a sourced row for MiniMax M3.

Knowledge

Reasoning

Coding

Agentic

Long context

Tool use

Performance

Human preference

Changelog
  1. Released MiniMax M3 — Terminal-Bench 2.0 66.0%, 1M context.
  2. MiniMax M2.7 retired as the flagship open-weights model.
Sources