DeepSeek V4.1 Flash 1017 vs MiMo V2.6 Pro: Benchmark Comparison
DeepSeek V4.1 Flash 1017 has the higher score on 5 of 9 shared benchmarks; MiMo V2.6 Pro leads on 4.
The largest observed score gap is 11.96 pts on Tax Agent Bench , where DeepSeek V4.1 Flash 1017 leads.
Reported ±1 standard-error ranges overlap on 8 of 9 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 1017 | MiMo V2.6 Pro | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 56.58% ±1.27 | 55.20% ±1.18 | 1.39 pts | Reported ±1 SE ranges overlap |
| Harvey's Legal Agent Benchmark | 9.17% ±2.51 | 10.83% ±2.45 | 1.67 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 45.19% ±3.46 | 47.12% ±3.47 | 1.92 pts | Reported ±1 SE ranges overlap |
| EMB | 61.80% ±3.19 | 62.86% ±3.14 | 1.05 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 56.96% ±2.05 | 57.34% ±0.57 | 0.38 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 44.35% ±3.57 | 32.39% ±3.36 | 11.96 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 44.21% ±4.46 | 43.01% ±4.32 | 1.20 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 31.82% ±2.62 | 31.31% ±3.07 | 0.51 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 87.73% ±2.75 | 85.22% ±3.39 | 2.51 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | DeepSeek V4.1 Flash 1017 average | MiMo V2.6 Pro average |
|---|---|---|
| Legal | 27.18% | 28.97% |
| Finance | 54.37% | 50.86% |
| Coding | 54.59% | 53.18% |
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 1017 cost | MiMo V2.6 Pro cost | DeepSeek V4.1 Flash 1017 latency | MiMo V2.6 Pro latency |
|---|---|---|---|---|
| Vals Index | $0.64 | $0.41 | 36m33s | 1h07m |
| Harvey's Legal Agent Benchmark | $0.54 | $0.22 | 21m39s | 25m53s |
| Legal Research Bench | $0.70 | $0.18 | 35m07s | 30m16s |
| EMB | $0.36 | $0.33 | 16m47s | 55m14s |
| Finance Agent (v2) | $0.46 | $0.20 | 14m39s | 10m48s |
| Tax Agent Bench | $0.46 | $0.13 | 21m04s | 25m48s |
| Code Migration | $1.86 | $0.69 | 2h11m | 2h49m |
| Terminal-Bench 4.0 | $0.54 | $0.50 | 41m10s | 2h42m |
| Vibe Code Bench v1.1 | $0.58 | $1.04 | 16m14s | 1h01m |
Results available only for DeepSeek V4.1 Flash 1017
None.
Results available only for MiMo V2.6 Pro
- MedCode
- MedScribe
- ProofBench v1.1
- MysteryMechanism
- Terminal-Bench-Science
- SAGE
- IOI
- ProgramBench
- SWE-bench Multimodal
- CyberBench v1.1
- SRE Bench
- Public Benefits Bench v1.1