IFBench ↗75.71%
MiniMax-M2.7 · IFBench · AA implementation ↗
MiniMax / MODEL EVIDENCE
MiniMax M2.7
30 benchmarks · 32 observations · 30 sources · released 2026-03-18
Scores at a glance
30 benchmarks · large dot: MiniMax M2.7; small dots: closest models where reported
Terminal-Bench 2.0 ↗45.06%
MiniMax-M2.7 · Terminal Bench · Accuracy mean · % ↗
1 more reported result
42.92% · MiniMax-M2.7 · Terminal Bench · Accuracy mean · % · Epoch AI · Terminal Bench ↗
WeirdML (v2) ↗36.95%
MiniMax-M2.7 · WeirdML · Accuracy · % ↗
MiniMax-M2.7 · GBAEval · Overall score · % ↗
CritPt ↗0.57%
MiniMax-M2.7 · CritPt · Accuracy · % ↗
ExploitBench ↗13.3%
MiniMax-M2.7 · ExploitBench · Mean capability · % ↗
MiniMax-M2.7 · ProofBench · Accuracy · % ↗
SciCode ↗50.12%
MiniMax-M2.7 · SciCode · Score · % ↗
Text Arena (Coding) ↗1397.52 Elo
MiniMax-M2.7 · WebDev Arena · Elo ↗
Epoch Capabilities Index (ECI) ↗145.78 ECI
MiniMax-M2.7 · General ECI · 26 October 2026 fit ↗
Legal Research Bench ↗10.58 %
MiniMax-M2.7 · Legal Research Bench · All-Pass ↗
LegalBench ↗83.98 %
MiniMax-M2.7 · LegalBench · Accuracy ↗
MiniMax-M2.7 · Harvey's Legal Agent Benchmark v1 · Accuracy ↗
CaseLaw v2 ↗60.88 %
MiniMax-M2.7 · CaseLaw · Accuracy ↗
Tax Agent Bench ↗6.87 %
MiniMax-M2.7 · Tax Agent Bench · Accuracy ↗
Vals Finance Agent v2 ↗27.89 %
MiniMax-M2.7 · Finance Agent v2 · Accuracy ↗
MiniMax-M2.7 · Excel Modeling Benchmark v1 · Accuracy ↗
CorpFin v2 ↗61.19 %
MiniMax-M2.7 · CorpFin · Accuracy ↗
MortgageTax & TaxEval v2 ↗66.56 %
MiniMax-M2.7 · TaxEval · Accuracy ↗
MedCode ↗34.44 %
MiniMax-M2.7 · MedCode · Accuracy ↗
MedScribe ↗79.87 %
MiniMax-M2.7 · MedScribe · Accuracy ↗
AIME 2024–2025 · Vals ↗91.04 %
MiniMax-M2.7 · AIME · Accuracy ↗
GPQA Diamond ↗86.62 %
MiniMax-M2.7 · GPQA Diamond · Accuracy ↗
MMLU-Pro ↗80.43 %
MiniMax-M2.7 · MMLU Pro · Accuracy ↗
Vibe Code Bench ↗11.93 %
MiniMax-M2.7 (OpenHands) · Vibe Code Bench · original task set ↗
MiniMax-M2.7 · ProgramBench · Accuracy ↗
Code Migration v1.1 ↗8.69 %
MiniMax-M2.7 · Code Migration v1 · Accuracy ↗
SWE-bench Verified ↗73.8 %
MiniMax-M2.7 (Mini-SWE-agent) · SWE-bench Verified · Score ↗
LiveCodeBench ↗79.93 %
MiniMax-M2.7 · LiveCodeBench · Accuracy ↗
ALE-Bench ↗599.25 Performance points
minimax-m2.7 · Average contest performance ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
13 of 20 shared benchmark comparisons
Claude Haiku 4.5 →Anthropic
17 of 20 shared benchmark comparisons
GPT-5.4 Mini →OpenAI
14 of 20 shared benchmark comparisons
How similarity is measured
Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.
Sources
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
- Arena leaderboard — WebDev evaluation ↗
- IFBench Benchmark | Artificial Analysis ↗
- SciCode Benchmark | Artificial Analysis ↗
- Ale Bench — methodology and published results ↗
- CritPt v1 ↗
- ExploitBench v0.1 ↗
- AI benchmarks & capabilities ↗
- MiniMax-M2.7 — GBA Eval ↗
- WeirdML ↗
- Terminal-Bench 2.0 leaderboard ↗
- AIME — evaluation methodology and results ↗
- CaseLaw — evaluation methodology and results ↗
- Code Migration v1 — evaluation methodology and results ↗
- CorpFin — evaluation methodology and results ↗
- Excel Modeling Benchmark v1 — evaluation methodology and results ↗
- Finance Agent v2 — evaluation methodology and results ↗
- GPQA Diamond — evaluation methodology and results ↗
- Harvey's Legal Agent Benchmark v1 — evaluation methodology and results ↗
- LiveCodeBench — evaluation methodology and results ↗
- LegalBench — evaluation methodology and results ↗
- Legal Research Bench — evaluation methodology and results ↗
- MedCode — evaluation methodology and results ↗
- MedScribe — evaluation methodology and results ↗
- MMLU Pro — evaluation methodology and results ↗
- ProgramBench — evaluation methodology and results ↗
- ProofBench — evaluation methodology and results ↗
- SWE-bench Verified — evaluation methodology and results ↗
- Tax Agent Bench — evaluation methodology and results ↗
- TaxEval — evaluation methodology and results ↗
- Vibe Code Bench — evaluation methodology and results ↗
Names used in source tables
MiniMax-M2.7 · MiniMax-M2.7 (Mini-SWE-agent) · MiniMax-M2.7 (OpenHands) · minimax-m2.7