MiniMax-M3 vs Inkling: Benchmark Comparison
MiniMax-M3 has the higher score on 19 of 24 shared benchmarks; Inkling leads on 5.
The largest observed score gap is 28.35 pts on Vibe Code Bench v1.1 , where MiniMax-M3 leads.
Reported ±1 standard-error ranges overlap on 6 of 24 shared benchmarks with comparable uncertainty data. This is not a pairwise statistical significance test.
Shared benchmark results
Scores and reported standard errors come from the latest visible Vals benchmark version available for both models.
| Benchmark | MiniMax-M3 | Inkling | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Finance Index | 48.37% ±1.41 | 38.25% ±1.51 | 10.12 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 29.81% ±3.18 | 28.36% ±3.13 | 1.44 pts | Reported ±1 SE ranges overlap |
| LegalBench | 85.42% ±0.44 | 82.95% ±0.50 | 2.47 pts | Reported ±1 SE ranges do not overlap |
| EMB v1.1 | 59.75% ±2.54 | 40.97% ±2.98 | 18.78 pts | Reported ±1 SE ranges do not overlap |
| Finance Agent v2.1 | 54.34% ±2.02 | 49.87% ±2.12 | 4.47 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 68.36% ±0.91 | 63.99% ±0.94 | 4.37 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 20.59% ±2.87 | 15.29% ±2.42 | 5.30 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 72.73% ±0.86 | 75.31% ±0.84 | 2.58 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 46.29% ±2.10 | 41.19% ±2.23 | 5.10 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 87.25% ±1.96 | 85.41% ±1.84 | 1.85 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 18.00% ±3.86 | 0.00% ±0.00 | 18.00 pts | Reported ±1 SE ranges do not overlap |
| GPQA Diamond | 92.68% ±1.44 | 87.12% ±1.78 | 5.55 pts | Reported ±1 SE ranges do not overlap |
| MMLU Pro | 84.22% ±0.36 | 86.30% ±0.34 | 2.08 pts | Reported ±1 SE ranges do not overlap |
| MMMU Pro | 81.16% ±0.94 | 78.50% ±0.99 | 2.66 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 50.57% ±3.44 | 36.55% ±3.22 | 14.02 pts | Reported ±1 SE ranges do not overlap |
| Code Migration v1.1 | 20.90% ±3.44 | 8.82% ±2.71 | 12.08 pts | Reported ±1 SE ranges do not overlap |
| LiveCodeBench | 82.15% ±1.05 | 85.52% ±1.00 | 3.36 pts | Reported ±1 SE ranges do not overlap |
| SkillsBench | 51.50% ±4.50 | 26.09% ±3.86 | 25.41 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench | 75.00% ±1.94 | 77.60% ±1.87 | 2.60 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 1.01% ±1.01 | 0.51% ±0.51 | 0.51 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench 1-100 | 9.22% ±2.86 | 7.33% ±1.84 | 1.89 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 47.57% ±5.44 | 19.21% ±3.66 | 28.35 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 65.00% ±6.10 | 66.79% ±5.95 | 1.79 pts | Reported ±1 SE ranges overlap |
| Public Benefits Bench v1.1 | 64.14% ±1.25 | 58.46% ±1.28 | 5.68 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | MiniMax-M3 average | Inkling average |
|---|---|---|
| Legal | 57.61% | 55.66% |
| Finance | 55.15% | 49.09% |
| Healthcare | 66.77% | 63.30% |
| Math | 18.00% | 0.00% |
| Academic | 86.02% | 83.97% |
| Education | 50.57% | 36.55% |
| Coding | 41.05% | 32.15% |
| Cyber | 65.00% | 66.79% |
| Social Mobility | 64.14% | 58.46% |
Cost and latency
Cost per test appears only where the benchmark reports it for both models. Latency is the measured completion time for that benchmark.
| Benchmark | MiniMax-M3 cost | Inkling cost | MiniMax-M3 latency | Inkling latency |
|---|---|---|---|---|
| Vals Finance Index | $1.22 | $0.97 | 10m01s | 16m29s |
| Legal Research Bench | $0.34 | $0.49 | 13m34s | 14m29s |
| LegalBench | N/A | N/A | 8.21s | 23.14s |
| EMB v1.1 | $3.18 | $1.67 | 21m39s | 26m22s |
| Finance Agent v2.1 | $0.30 | $0.89 | 3m09s | 9m34s |
| MortgageTax | N/A | N/A | 26.76s | 69.11s |
| Tax Agent Bench | $0.16 | $0.35 | 5m15s | 13m31s |
| TaxEval v2 | N/A | N/A | 94.04s | 3m18s |
| MedCode | N/A | N/A | 63.12s | 2m47s |
| MedScribe | N/A | N/A | 2m04s | 4m05s |
| ProofBench v1.1 | $0.42 | $0.20 | 11m59s | 7m16s |
| GPQA Diamond | N/A | N/A | 4m39s | 5m09s |
| MMLU Pro | N/A | N/A | 41.39s | 73.26s |
| MMMU Pro | N/A | N/A | 68.35s | 2m04s |
| SAGE | N/A | N/A | 118.54s | 3m18s |
| Code Migration v1.1 | $8.39 | $6.13 | 1h26m | 24m11s |
| LiveCodeBench | N/A | N/A | 6m07s | 4m18s |
| SkillsBench | $0.50 | $0.83 | 15m28s | 15m39s |
| SWE-bench | $0.42 | $0.59 | 12m06s | 6m34s |
| Terminal-Bench 4.0 | $7.23 | $2.50 | 1h16m | 22m59s |
| Vibe Code Bench 1-100 | $3.03 | $5.00 | 26m10s | 50m02s |
| Vibe Code Bench v1.1 | $6.45 | $1.61 | 1h25m | 18m25s |
| CyberBench v1.1 | $3.41 | $1.46 | 40m56s | 13m31s |
| Public Benefits Bench v1.1 | $0.25 | $0.37 | 11m31s | 13m31s |
Results available only for MiniMax-M3
- Vals Inference Provider Index
- SWE-bench Multimodal
Results available only for Inkling
- Vals Index
- Vals Legal Index
- HLAB-Vals-1.1
- IOI
- ProgramBench
- SRE Bench