Qwen 3.7 Max vs MiMo V2.5: Benchmark Comparison
Qwen 3.7 Max has the higher score on 12 of 15 shared benchmarks; MiMo V2.5 leads on 2.
The largest observed score gap is 18.24 pts on Tax Agent Bench , where Qwen 3.7 Max leads.
Reported ±1 standard-error ranges overlap on 5 of 15 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 | Qwen 3.7 Max | MiMo V2.5 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | 1.67% ±0.83 | 1.67% ±0.00 | 0.00 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 25.48% ±3.03 | 9.13% ±2.00 | 16.35 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 84.91% ±0.41 | 78.89% ±0.51 | 6.03 pts | Reported ±1 SE ranges do not overlap |
| EMB | 57.04% ±2.85 | 55.09% ±3.13 | 1.95 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 47.78% ±0.73 | 36.73% ±0.11 | 11.05 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 47.41% ±3.24 | 29.16% ±2.71 | 18.24 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 75.31% ±0.84 | 71.83% ±0.88 | 3.47 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 38.75% ±2.20 | 31.89% ±2.02 | 6.86 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 79.40% ±1.91 | 72.15% ±1.85 | 7.25 pts | Reported ±1 SE ranges do not overlap |
| GPQA Diamond | 90.15% ±1.50 | 81.57% ±2.05 | 8.59 pts | Reported ±1 SE ranges do not overlap |
| MMLU Pro | 89.31% ±0.30 | 82.93% ±0.37 | 6.38 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 13.07% ±2.86 | 14.23% ±3.69 | 1.16 pts | Reported ±1 SE ranges overlap |
| LiveCodeBench | 87.06% ±0.97 | 81.51% ±1.07 | 5.55 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench | 68.80% ±2.07 | 71.00% ±2.03 | 2.20 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 47.67% ±4.63 | 42.17% ±4.57 | 5.50 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Qwen 3.7 Max average | MiMo V2.5 average |
|---|---|---|
| Legal | 37.35% | 29.90% |
| Finance | 56.88% | 48.20% |
| Healthcare | 59.07% | 52.02% |
| Academic | 89.73% | 82.25% |
| Coding | 54.15% | 52.23% |
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 | Qwen 3.7 Max cost | MiMo V2.5 cost | Qwen 3.7 Max latency | MiMo V2.5 latency |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | $1.58 | $0.04 | 32m03s | 6m59s |
| Legal Research Bench | $1.40 | $0.03 | 13m29s | 6m31s |
| LegalBench | N/A | N/A | 11.85s | 11.18s |
| EMB | $2.86 | $0.08 | 32m37s | 30m38s |
| Finance Agent (v2) | $0.55 | $0.09 | 10m35s | 5m39s |
| Tax Agent Bench | $0.44 | $0.02 | 3m30s | 8m21s |
| TaxEval v2 | N/A | N/A | 57.68s | 26.30s |
| MedCode | N/A | N/A | 28.75s | 17.70s |
| MedScribe | N/A | N/A | 108.05s | 20.26s |
| GPQA Diamond | N/A | N/A | 2m07s | 89.39s |
| MMLU Pro | N/A | N/A | 29.15s | 30.25s |
| Code Migration | $1.93 | $0.10 | 25m32s | 37m55s |
| LiveCodeBench | N/A | N/A | 3m12s | 108.61s |
| SWE-bench | $3.23 | $0.01 | 16m16s | 4m04s |
| Vibe Code Bench v1.1 | $11.40 | $0.07 | 1h18m | 26m20s |
Results available only for Qwen 3.7 Max
None.
Results available only for MiMo V2.5
- MortgageTax
- ProofBench v1.1
- MMMU Pro
- SAGE