Gemini 3.8 Flash vs MiMo V2.6 Pro: Benchmark Comparison
Gemini 3.8 Flash has the higher score on 9 of 23 shared benchmarks; MiMo V2.6 Pro leads on 14.
The largest observed score gap is 29.11 pts on CyberBench v1.1 , where MiMo V2.6 Pro leads.
Reported ±1 standard-error ranges overlap on 11 of 23 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 | Gemini 3.8 Flash | MiMo V2.6 Pro | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Finance Index | 60.46% ±1.45 | 57.85% ±1.57 | 2.61 pts | Reported ±1 SE ranges overlap |
| Vals Index | 55.95% ±1.14 | 57.81% ±1.24 | 1.86 pts | Reported ±1 SE ranges overlap |
| Vals Legal Index | 69.78% ±1.78 | 72.14% ±1.83 | 2.36 pts | Reported ±1 SE ranges overlap |
| HLAB-Vals-1.1 | 12.08% ±2.58 | 13.75% ±2.84 | 1.67 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 38.94% ±3.39 | 47.12% ±3.47 | 8.17 pts | Reported ±1 SE ranges do not overlap |
| EMB v1.1 | 72.87% ±2.42 | 67.35% ±2.90 | 5.52 pts | Reported ±1 SE ranges do not overlap |
| Finance Agent v2.1 | 65.76% ±2.04 | 64.26% ±2.02 | 1.50 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 32.12% ±3.37 | 32.39% ±3.36 | 0.26 pts | Reported ±1 SE ranges overlap |
| MedCode | 48.13% ±2.18 | 44.97% ±2.10 | 3.17 pts | Reported ±1 SE ranges overlap |
| MedScribe | 84.50% ±1.94 | 88.31% ±1.94 | 3.81 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 48.00% ±5.02 | 70.00% ±4.61 | 22.00 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 36.49% ±3.24 | 15.31% ±2.42 | 21.17 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench-Science | 8.57% ±3.37 | 2.86% ±2.01 | 5.71 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 35.06% ±3.36 | 45.05% ±3.40 | 9.99 pts | Reported ±1 SE ranges do not overlap |
| Code Migration v1.1 | 38.65% ±3.65 | 48.79% ±3.88 | 10.14 pts | Reported ±1 SE ranges do not overlap |
| IOI | 56.94% ±2.26 | 39.33% ±2.41 | 17.61 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 1.00% ±0.70 | 0.50% ±0.50 | 0.50 pts | Reported ±1 SE ranges overlap |
| SWE-bench Multimodal | 40.83% ±2.25 | 41.25% ±2.25 | 0.42 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 19.70% ±2.31 | 31.31% ±3.07 | 11.62 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 78.65% ±3.88 | 85.22% ±3.39 | 6.57 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 43.75% ±2.21 | 72.86% ±5.50 | 29.11 pts | Reported ±1 SE ranges do not overlap |
| SRE Bench | 11.07% ±1.94 | 3.05% ±1.06 | 8.02 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 65.29% ±1.24 | 68.94% ±1.20 | 3.65 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Gemini 3.8 Flash average | MiMo V2.6 Pro average |
|---|---|---|
| Legal | 25.51% | 30.43% |
| Finance | 56.92% | 54.66% |
| Healthcare | 66.32% | 66.64% |
| Math | 48.00% | 70.00% |
| Science | 22.53% | 9.09% |
| Education | 35.06% | 45.05% |
| Coding | 39.30% | 41.07% |
| Cyber | 27.41% | 37.95% |
| Social Mobility | 65.29% | 68.94% |
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 | Gemini 3.8 Flash cost | MiMo V2.6 Pro cost | Gemini 3.8 Flash latency | MiMo V2.6 Pro latency |
|---|---|---|---|---|
| Vals Finance Index | $3.70 | $0.22 | 7m44s | 30m48s |
| Vals Index | $6.06 | $0.43 | 35m58s | 1h13m |
| Vals Legal Index | $2.05 | $0.18 | 9m15s | 29m02s |
| HLAB-Vals-1.1 | $3.66 | $0.22 | 19m22s | 31m02s |
| Legal Research Bench | $1.63 | $0.18 | 5m24s | 30m16s |
| EMB v1.1 | $8.24 | $0.33 | 12m49s | 55m14s |
| Finance Agent v2.1 | $1.99 | $0.18 | 7m24s | 11m22s |
| Tax Agent Bench | $0.86 | $0.13 | 2m58s | 25m48s |
| MedCode | N/A | N/A | 42.89s | 2m57s |
| MedScribe | N/A | N/A | 21.14s | 2m53s |
| ProofBench v1.1 | $0.60 | $0.25 | 6m09s | 1h02m |
| MysteryMechanism | $1.73 | $0.15 | 5m22s | 44m00s |
| Terminal-Bench-Science | $5.64 | $0.62 | 55m11s | 3h12m |
| SAGE | N/A | N/A | 28.64s | 3m40s |
| Code Migration v1.1 | $15.79 | $0.99 | 2h35m | 3h28m |
| IOI | $3.98 | $0.84 | 14m20s | 2h26m |
| ProgramBench | $10.93 | $0.73 | 46m39s | 3h00m |
| SWE-bench Multimodal | $3.69 | $0.12 | 22m18s | 46m03s |
| Terminal-Bench 4.0 | $8.65 | $0.50 | 1h47m | 2h42m |
| Vibe Code Bench v1.1 | $6.87 | $1.04 | 8m39s | 1h01m |
| CyberBench v1.1 | $1.31 | $0.09 | 9m49s | 27m41s |
| SRE Bench | $27.46 | $0.53 | 2h28m | 2h23m |
| Public Benefits Bench v1.1 | $1.01 | $0.08 | 3m21s | 37m14s |
Results available only for Gemini 3.8 Flash
- Vals RSI Index
- Vals Science Index
- LegalBench
- MortgageTax
- TaxEval v2
- BioMysteryBench
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- LiveCodeBench
- SkillsBench
- SWE-bench
- Vibe Code Bench 1-100
- CUA-bench
- Time Horizon Index: KSP
Results available only for MiMo V2.6 Pro
None.