DeepSeek V4 Pro 0813 vs Mistral Large 4: Benchmark Comparison
DeepSeek V4 Pro 0813 has the higher score on 7 of 14 shared benchmarks; Mistral Large 4 leads on 7.
The largest observed score gap is 40.00 pts on ProofBench v1.1 , where DeepSeek V4 Pro 0813 leads.
Reported ±1 standard-error ranges overlap on 6 of 14 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 | DeepSeek V4 Pro 0813 | Mistral Large 4 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Finance Index | 52.94% ±1.41 | 56.90% ±1.42 | 3.96 pts | Reported ±1 SE ranges do not overlap |
| Vals Legal Index | 69.01% ±1.77 | 65.61% ±1.63 | 3.40 pts | Reported ±1 SE ranges do not overlap |
| HLAB-Vals-1.1 | 10.83% ±2.72 | 17.50% ±3.22 | 6.67 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 40.87% ±3.42 | 31.73% ±3.23 | 9.13 pts | Reported ±1 SE ranges do not overlap |
| EMB v1.1 | 65.13% ±2.35 | 68.97% ±2.34 | 3.84 pts | Reported ±1 SE ranges overlap |
| Finance Agent v2.1 | 57.12% ±2.08 | 61.38% ±2.02 | 4.27 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 26.76% ±3.09 | 30.39% ±3.31 | 3.63 pts | Reported ±1 SE ranges overlap |
| MedCode | 42.47% ±2.16 | 40.69% ±2.17 | 1.78 pts | Reported ±1 SE ranges overlap |
| MedScribe | 80.17% ±2.00 | 80.43% ±2.00 | 0.26 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 50.00% ±5.03 | 10.00% ±3.02 | 40.00 pts | Reported ±1 SE ranges do not overlap |
| IOI | 51.61% ±2.40 | 45.28% ±3.78 | 6.33 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 14.14% ±2.02 | 27.78% ±1.82 | 13.64 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench 1-100 | 17.53% ±3.41 | 14.26% ±3.21 | 3.27 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 82.30% ±3.15 | 78.40% ±3.55 | 3.89 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | DeepSeek V4 Pro 0813 average | Mistral Large 4 average |
|---|---|---|
| Legal | 25.85% | 24.62% |
| Finance | 49.67% | 53.58% |
| Healthcare | 61.32% | 60.56% |
| Math | 50.00% | 10.00% |
| Coding | 41.40% | 41.43% |
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 | DeepSeek V4 Pro 0813 cost | Mistral Large 4 cost | DeepSeek V4 Pro 0813 latency | Mistral Large 4 latency |
|---|---|---|---|---|
| Vals Finance Index | $0.88 | $2.86 | 24m00s | 51m50s |
| Vals Legal Index | $0.87 | $4.08 | 26m33s | 1h09m |
| HLAB-Vals-1.1 | $0.80 | $4.18 | 10m55s | 58m04s |
| Legal Research Bench | $1.09 | $5.33 | 42m07s | 1h33m |
| EMB v1.1 | $1.08 | $4.84 | 37m27s | 1h22m |
| Finance Agent v2.1 | $0.82 | $1.02 | 7m57s | 16m26s |
| Tax Agent Bench | $0.73 | $2.72 | 26m37s | 57m18s |
| MedCode | N/A | N/A | 3m01s | 68.68s |
| MedScribe | N/A | N/A | 2m36s | 2m05s |
| ProofBench v1.1 | $0.06 | $0.64 | 12m12s | 20m59s |
| IOI | $2.15 | $5.39 | 1h07m | 1h24m |
| Terminal-Bench 4.0 | $3.31 | $10.25 | 1h32m | 1h40m |
| Vibe Code Bench 1-100 | $5.89 | $6.75 | 37m42s | 59m53s |
| Vibe Code Bench v1.1 | $0.36 | $39.46 | 1h06m | 2h53m |
Results available only for DeepSeek V4 Pro 0813
- Vals Index
- LegalBench
- TaxEval v2
- Terminal-Bench-Science
- GPQA Diamond
- MMLU Pro
- Code Migration v1.1
- LiveCodeBench
- ProgramBench
- SkillsBench
- SWE-bench
- CyberBench v1.1
- SRE Bench
Results available only for Mistral Large 4
- BioMysteryBench
- MysteryMechanism
- SWE-bench Multimodal