Qwen 3.7 Max vs Gemini 3.1 Pro Preview (02/26): Benchmark Comparison
Qwen 3.7 Max has the higher score on 8 of 15 shared benchmarks; Gemini 3.1 Pro Preview (02/26) leads on 7.
The largest observed score gap is 20.31 pts on MedCode , where Gemini 3.1 Pro Preview (02/26) 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 | Gemini 3.1 Pro Preview (02/26) | Gap | Reported uncertainty |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | 1.67% ±0.83 | 0.00% ±0.00 | 1.67 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 25.48% ±3.03 | 20.67% ±2.81 | 4.81 pts | Reported ±1 SE ranges overlap |
| LegalBench | 84.91% ±0.41 | 87.40% ±0.33 | 2.48 pts | Reported ±1 SE ranges do not overlap |
| EMB | 57.04% ±2.85 | 52.62% ±2.97 | 4.42 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 47.78% ±0.73 | 42.98% ±1.21 | 4.79 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 47.41% ±3.24 | 37.84% ±3.10 | 9.56 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 75.31% ±0.84 | 72.88% ±0.86 | 2.42 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 38.75% ±2.20 | 59.06% ±2.00 | 20.31 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 79.40% ±1.91 | 76.11% ±1.92 | 3.28 pts | Reported ±1 SE ranges overlap |
| GPQA Diamond | 90.15% ±1.50 | 95.45% ±1.05 | 5.30 pts | Reported ±1 SE ranges do not overlap |
| MMLU Pro | 89.31% ±0.30 | 90.99% ±0.28 | 1.68 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 13.07% ±2.86 | 17.31% ±3.93 | 4.24 pts | Reported ±1 SE ranges overlap |
| LiveCodeBench | 87.06% ±0.97 | 88.48% ±0.93 | 1.43 pts | Reported ±1 SE ranges overlap |
| SWE-bench | 68.80% ±2.07 | 78.80% ±1.83 | 10.00 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 47.67% ±4.63 | 32.03% ±4.34 | 15.64 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Qwen 3.7 Max average | Gemini 3.1 Pro Preview (02/26) average |
|---|---|---|
| Legal | 37.35% | 36.02% |
| Finance | 56.88% | 51.58% |
| Healthcare | 59.07% | 67.59% |
| Academic | 89.73% | 93.22% |
| Coding | 54.15% | 54.16% |
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 | Gemini 3.1 Pro Preview (02/26) cost | Qwen 3.7 Max latency | Gemini 3.1 Pro Preview (02/26) latency |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | $1.58 | $1.15 | 32m03s | 7m17s |
| Legal Research Bench | $1.40 | $1.10 | 13m29s | 4m02s |
| LegalBench | N/A | N/A | 11.85s | 10.06s |
| EMB | $2.86 | $3.77 | 32m37s | 11m49s |
| Finance Agent (v2) | $0.55 | $1.59 | 10m35s | 3m28s |
| Tax Agent Bench | $0.44 | $0.30 | 3m30s | 100.96s |
| TaxEval v2 | N/A | N/A | 57.68s | 41.93s |
| MedCode | N/A | N/A | 28.75s | 38.52s |
| MedScribe | N/A | N/A | 108.05s | 69.14s |
| GPQA Diamond | N/A | N/A | 2m07s | 65.76s |
| MMLU Pro | N/A | N/A | 29.15s | 23.91s |
| Code Migration | $1.93 | $1.61 | 25m32s | 12m01s |
| LiveCodeBench | N/A | N/A | 3m12s | 88.89s |
| SWE-bench | $3.23 | $0.78 | 16m16s | 5m12s |
| Vibe Code Bench v1.1 | $11.40 | $3.83 | 1h18m | 20m12s |
Results available only for Qwen 3.7 Max
None.
Results available only for Gemini 3.1 Pro Preview (02/26)
- Vals Index
- MortgageTax
- ProofBench v1.1
- BioMysteryBench
- MysteryMechanism
- Terminal-Bench Science
- MMMU Pro
- SAGE
- IOI
- ProgramBench
- Terminal-Bench 4.0
- Vibe Code Bench 1-100
- Public Benefits Bench v1.1