Qwen 3.7 Plus vs DeepSeek V4: Benchmark Comparison
Qwen 3.7 Plus has the higher score on 1 of 8 shared benchmarks; DeepSeek V4 leads on 7.
The largest observed score gap is 19.79 pts on Tax Agent Bench , where DeepSeek V4 leads.
Reported ±1 standard-error ranges overlap on 3 of 8 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 Plus | DeepSeek V4 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | 0.00% ±0.00 | 3.75% ±1.43 | 3.75 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 16.35% ±2.57 | 23.08% ±2.93 | 6.73 pts | Reported ±1 SE ranges do not overlap |
| EMB | 49.34% ±3.18 | 51.62% ±3.05 | 2.28 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 38.22% ±1.04 | 44.08% ±0.65 | 5.86 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 38.71% ±2.81 | 58.50% ±3.13 | 19.79 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 12.86% ±2.93 | 26.20% ±4.04 | 13.35 pts | Reported ±1 SE ranges do not overlap |
| SkillsBench | 54.30% ±4.33 | 51.27% ±4.60 | 3.02 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 46.39% ±4.61 | 49.93% ±4.77 | 3.54 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Qwen 3.7 Plus average | DeepSeek V4 average |
|---|---|---|
| Legal | 8.17% | 13.41% |
| Finance | 42.09% | 51.40% |
| Coding | 37.85% | 42.47% |
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 Plus cost | DeepSeek V4 cost | Qwen 3.7 Plus latency | DeepSeek V4 latency |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | $0.23 | $0.74 | 8m49s | 6m42s |
| Legal Research Bench | $0.30 | $0.68 | 13m52s | 17m05s |
| EMB | $0.60 | $0.80 | 36m49s | 13m42s |
| Finance Agent (v2) | $0.36 | $0.88 | 9m02s | 25m26s |
| Tax Agent Bench | $0.09 | $0.72 | 5m10s | 20m45s |
| Code Migration | $0.42 | $1.07 | 35m47s | 28m58s |
| SkillsBench | $0.14 | $0.40 | 12m04s | 8m16s |
| Vibe Code Bench v1.1 | $1.08 | $2.21 | 37m12s | 56m59s |
Results available only for Qwen 3.7 Plus
- MortgageTax
- SAGE
Results available only for DeepSeek V4
- Vals Index
- LegalBench
- TaxEval v2
- MedCode
- MedScribe
- ProofBench v1.1
- GPQA Diamond
- MMLU Pro
- LiveCodeBench
- ProgramBench
- SWE-bench
- Terminal-Bench 4.0
- Public Benefits Bench v1.1