DeepSeek V4.1 Flash vs MiMo V2.6 Pro: Benchmark Comparison
DeepSeek V4.1 Flash has the higher score on 5 of 20 shared benchmarks; MiMo V2.6 Pro leads on 14.
The largest observed score gap is 16.00 pts on ProofBench v1.1 , where MiMo V2.6 Pro leads.
Reported ±1 standard-error ranges overlap on 12 of 20 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 V4.1 Flash | MiMo V2.6 Pro | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 51.32% ±1.13 | 55.20% ±1.18 | 3.88 pts | Reported ±1 SE ranges do not overlap |
| Harvey's Legal Agent Benchmark | 6.67% ±1.83 | 10.83% ±2.45 | 4.17 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 41.35% ±3.42 | 47.12% ±3.47 | 5.77 pts | Reported ±1 SE ranges overlap |
| EMB | 57.21% ±3.23 | 62.86% ±3.14 | 5.65 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 53.48% ±0.39 | 57.34% ±0.57 | 3.86 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 62.46% ±3.15 | 64.94% ±3.21 | 2.48 pts | Reported ±1 SE ranges overlap |
| MedCode | 41.17% ±2.04 | 44.97% ±2.10 | 3.80 pts | Reported ±1 SE ranges overlap |
| MedScribe | 85.50% ±1.92 | 88.31% ±1.94 | 2.81 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 54.00% ±5.01 | 70.00% ±4.61 | 16.00 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 21.17% ±2.75 | 15.31% ±2.42 | 5.86 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench Science | 0.00% ±0.00 | 2.86% ±2.01 | 2.86 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 47.88% ±3.44 | 45.05% ±3.40 | 2.83 pts | Reported ±1 SE ranges overlap |
| Code Migration | 45.62% ±4.29 | 43.01% ±4.32 | 2.61 pts | Reported ±1 SE ranges overlap |
| IOI | 40.28% ±2.56 | 39.33% ±2.41 | 0.95 pts | Reported ±1 SE ranges overlap |
| ProgramBench | 0.50% ±0.50 | 0.50% ±0.50 | 0.00 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 19.70% ±1.75 | 31.31% ±3.07 | 11.62 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 84.74% ±2.89 | 85.22% ±3.39 | 0.48 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 73.69% ±5.48 | 72.86% ±5.50 | 0.83 pts | Reported ±1 SE ranges overlap |
| SRE Bench | 0.76% ±0.54 | 3.05% ±1.06 | 2.29 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 64.28% ±1.25 | 68.94% ±1.20 | 4.67 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | DeepSeek V4.1 Flash average | MiMo V2.6 Pro average |
|---|---|---|
| Legal | 24.01% | 28.97% |
| Finance | 57.72% | 61.71% |
| Healthcare | 63.34% | 66.64% |
| Math | 54.00% | 70.00% |
| Science | 10.59% | 9.09% |
| Education | 47.88% | 45.05% |
| Coding | 38.17% | 39.88% |
| Cyber | 37.23% | 37.95% |
| Social Mobility | 64.28% | 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 | DeepSeek V4.1 Flash cost | MiMo V2.6 Pro cost | DeepSeek V4.1 Flash latency | MiMo V2.6 Pro latency |
|---|---|---|---|---|
| Vals Index | $0.33 | $0.41 | 27m53s | 1h07m |
| Harvey's Legal Agent Benchmark | $0.17 | $0.22 | 9m39s | 25m49s |
| Legal Research Bench | $0.25 | $0.18 | 11m42s | 30m16s |
| EMB | $0.23 | $0.33 | 13m00s | 55m14s |
| Finance Agent (v2) | $0.21 | $0.20 | 6m04s | 10m48s |
| Tax Agent Bench | $0.13 | $0.13 | 5m49s | 25m48s |
| MedCode | N/A | N/A | 35.13s | 2m57s |
| MedScribe | N/A | N/A | 53.83s | 2m53s |
| ProofBench v1.1 | $0.13 | $0.25 | 9m07s | 1h02m |
| MysteryMechanism | $0.16 | $0.15 | 8m04s | 44m00s |
| Terminal-Bench Science | $0.47 | $0.62 | 2h57m | 3h12m |
| SAGE | N/A | N/A | 35.78s | 3m40s |
| Code Migration | $0.94 | $0.69 | 1h45m | 2h49m |
| IOI | $0.24 | $0.84 | 18m30s | 2h26m |
| ProgramBench | $0.90 | $0.73 | 1h02m | 3h00m |
| Terminal-Bench 4.0 | $0.50 | $0.50 | 48m22s | 2h42m |
| Vibe Code Bench v1.1 | $0.41 | $1.04 | 15m28s | 1h01m |
| CyberBench v1.1 | $0.07 | $0.09 | 13m50s | 27m41s |
| SRE Bench | $0.55 | $0.53 | 1h01m | 2h23m |
| Public Benefits Bench v1.1 | $0.07 | $0.08 | 9m59s | 37m14s |
Results available only for DeepSeek V4.1 Flash
- LegalBench
- BioMysteryBench
- SkillsBench
- Vibe Code Bench 1-100
Results available only for MiMo V2.6 Pro
None.