Mistral Large 4 vs Hy4 Preview: Benchmark Comparison
Mistral Large 4 has the higher score on 4 of 14 shared benchmarks; Hy4 Preview leads on 10.
The largest observed score gap is 65.00 pts on ProofBench v1.1 , where Hy4 Preview leads.
Reported ±1 standard-error ranges overlap on 8 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 | Mistral Large 4 | Hy4 Preview | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Finance Index | 56.90% ±1.42 | 57.51% ±1.45 | 0.61 pts | Reported ±1 SE ranges overlap |
| Vals Legal Index | 65.61% ±1.63 | 71.70% ±1.84 | 6.09 pts | Reported ±1 SE ranges do not overlap |
| HLAB-Vals-1.1 | 17.50% ±3.22 | 8.75% ±2.19 | 8.75 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 31.73% ±3.23 | 45.19% ±3.46 | 13.46 pts | Reported ±1 SE ranges do not overlap |
| EMB v1.1 | 68.97% ±2.34 | 68.95% ±2.40 | 0.03 pts | Reported ±1 SE ranges overlap |
| Finance Agent v2.1 | 61.38% ±2.02 | 61.62% ±2.06 | 0.24 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 30.39% ±3.31 | 32.62% ±3.38 | 2.23 pts | Reported ±1 SE ranges overlap |
| MedCode | 40.69% ±2.17 | 43.25% ±2.13 | 2.56 pts | Reported ±1 SE ranges overlap |
| MedScribe | 80.43% ±2.00 | 83.60% ±2.06 | 3.17 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 10.00% ±3.02 | 75.00% ±4.35 | 65.00 pts | Reported ±1 SE ranges do not overlap |
| BioMysteryBench | 67.04% ±2.67 | 69.26% ±2.59 | 2.22 pts | Reported ±1 SE ranges overlap |
| IOI | 45.28% ±3.78 | 59.33% ±4.60 | 14.05 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 27.78% ±1.82 | 8.08% ±1.34 | 19.70 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 78.40% ±3.55 | 77.48% ±4.04 | 0.93 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Mistral Large 4 average | Hy4 Preview average |
|---|---|---|
| Legal | 24.62% | 26.97% |
| Finance | 53.58% | 54.40% |
| Healthcare | 60.56% | 63.42% |
| Math | 10.00% | 75.00% |
| Science | 67.04% | 69.26% |
| Coding | 50.49% | 48.30% |
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 | Mistral Large 4 cost | Hy4 Preview cost | Mistral Large 4 latency | Hy4 Preview latency |
|---|---|---|---|---|
| Vals Finance Index | $2.86 | $0.68 | 51m50s | 32m15s |
| Vals Legal Index | $4.08 | $0.81 | 1h09m | 46m04s |
| HLAB-Vals-1.1 | $4.18 | $0.98 | 58m04s | 33m09s |
| Legal Research Bench | $5.33 | $0.85 | 1h33m | 1h04m |
| EMB v1.1 | $4.84 | $0.79 | 1h22m | 30m39s |
| Finance Agent v2.1 | $1.02 | $0.67 | 16m26s | 25m03s |
| Tax Agent Bench | $2.72 | $0.59 | 57m18s | 41m04s |
| MedCode | N/A | N/A | 68.68s | 6m39s |
| MedScribe | N/A | N/A | 2m05s | 5m50s |
| ProofBench v1.1 | $0.64 | $0.34 | 20m59s | 22m54s |
| BioMysteryBench | $2.07 | $0.36 | 25m32s | 23m30s |
| IOI | $5.39 | $1.35 | 1h24m | 1h01m |
| Terminal-Bench 4.0 | $10.25 | $2.10 | 1h40m | 2h10m |
| Vibe Code Bench v1.1 | $39.46 | $2.18 | 2h53m | 41m33s |
Results available only for Mistral Large 4
- MysteryMechanism
- SWE-bench Multimodal
- Vibe Code Bench 1-100
Results available only for Hy4 Preview
- Vals Index
- LegalBench
- Terminal-Bench-Science
- Code Migration v1.1
- ProgramBench
- CyberBench v1.1
- SRE Bench
- Public Benefits Bench v1.1