GPT 5.4 Nano vs MiMo V2.5 Pro: Benchmark Comparison
GPT 5.4 Nano has the higher score on 3 of 15 shared benchmarks; MiMo V2.5 Pro leads on 12.
The largest observed score gap is 10.48 pts on EMB , where MiMo V2.5 Pro leads.
Reported ±1 standard-error ranges overlap on 3 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 | GPT 5.4 Nano | MiMo V2.5 Pro | Gap | Reported uncertainty |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | 0.00% ±0.00 | 2.08% ±0.83 | 2.08 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 6.25% ±1.68 | 15.87% ±2.54 | 9.62 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 77.92% ±0.43 | 77.13% ±0.55 | 0.79 pts | Reported ±1 SE ranges overlap |
| EMB | 44.75% ±3.12 | 55.23% ±3.11 | 10.48 pts | Reported ±1 SE ranges do not overlap |
| Finance Agent (v2) | 38.22% ±1.19 | 41.50% ±1.19 | 3.29 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 26.58% ±2.52 | 34.79% ±2.62 | 8.21 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 67.42% ±0.93 | 73.79% ±0.86 | 6.38 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 41.03% ±2.26 | 32.48% ±1.91 | 8.55 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 77.09% ±1.89 | 83.73% ±2.06 | 6.64 pts | Reported ±1 SE ranges do not overlap |
| GPQA Diamond | 77.53% ±2.42 | 82.58% ±2.47 | 5.05 pts | Reported ±1 SE ranges do not overlap |
| MMLU Pro | 77.17% ±0.43 | 84.59% ±0.45 | 7.42 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 14.47% ±4.03 | 21.56% ±4.13 | 7.09 pts | Reported ±1 SE ranges overlap |
| LiveCodeBench | 84.01% ±1.04 | 81.35% ±1.07 | 2.66 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench | 69.80% ±2.06 | 74.00% ±1.96 | 4.20 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 26.10% ±5.08 | 34.11% ±4.53 | 8.02 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | GPT 5.4 Nano average | MiMo V2.5 Pro average |
|---|---|---|
| Legal | 28.06% | 31.69% |
| Finance | 44.24% | 51.33% |
| Healthcare | 59.06% | 58.11% |
| Academic | 77.35% | 83.59% |
| Coding | 48.59% | 52.76% |
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 | GPT 5.4 Nano cost | MiMo V2.5 Pro cost | GPT 5.4 Nano latency | MiMo V2.5 Pro latency |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | $0.18 | $0.11 | 6m03s | 9m04s |
| Legal Research Bench | $0.14 | $0.08 | 8m47s | 6m56s |
| LegalBench | N/A | N/A | 2.70s | 27.14s |
| EMB | $0.57 | $0.22 | 29m27s | 29m08s |
| Finance Agent (v2) | $0.16 | $0.21 | 5m35s | 8m17s |
| Tax Agent Bench | $0.07 | $0.04 | 3m40s | 6m17s |
| TaxEval v2 | N/A | N/A | 15.90s | 62.83s |
| MedCode | N/A | N/A | 8.04s | 38.45s |
| MedScribe | N/A | N/A | 20.53s | 90.71s |
| GPQA Diamond | N/A | N/A | 29.50s | 4m11s |
| MMLU Pro | N/A | N/A | 8.21s | 67.76s |
| Code Migration | $0.42 | $0.24 | 26m34s | 48m53s |
| LiveCodeBench | N/A | N/A | 60.09s | 4m31s |
| SWE-bench | $0.10 | $0.02 | 4m23s | 7m51s |
| Vibe Code Bench v1.1 | $1.28 | $0.17 | 54m36s | 40m04s |
Results available only for GPT 5.4 Nano
- MortgageTax
- MMMU Pro
- SAGE
Results available only for MiMo V2.5 Pro
- Vals Index
- ProofBench v1.1
- Terminal-Bench 4.0