Model comparison

Claude Sonnet 5.5 vs MiMo V2.6 Flash: Benchmark Comparison

Claude Sonnet 5.5 has the higher score on 17 of 20 shared benchmarks; MiMo V2.6 Flash leads on 3.

The largest observed score gap is 37.00 pts on ProofBench v1.1 , where Claude Sonnet 5.5 leads.

Reported ±1 standard-error ranges overlap on 1 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 Claude Sonnet 5.5 MiMo V2.6 Flash Gap Reported uncertainty
Vals Index 69.22% ±0.96 59.58% ±1.13 9.63 pts Reported ±1 SE ranges do not overlap
Harvey's Legal Agent Benchmark 2.92% ±1.36 11.25% ±2.29 8.33 pts Reported ±1 SE ranges do not overlap
Legal Research Bench 48.08% ±3.47 37.98% ±3.37 10.10 pts Reported ±1 SE ranges do not overlap
EMB 75.71% ±2.44 65.46% ±2.71 10.25 pts Reported ±1 SE ranges do not overlap
Finance Agent (v2) 58.10% ±0.67 56.28% ±0.46 1.83 pts Reported ±1 SE ranges do not overlap
Tax Agent Bench 73.39% ±2.98 59.90% ±3.13 13.49 pts Reported ±1 SE ranges do not overlap
MedCode 52.92% ±2.12 41.06% ±2.03 11.86 pts Reported ±1 SE ranges do not overlap
MedScribe 91.10% ±1.96 85.28% ±1.99 5.83 pts Reported ±1 SE ranges do not overlap
ProofBench v1.1 100.00% ±0.00 63.00% ±4.85 37.00 pts Reported ±1 SE ranges do not overlap
BioMysteryBench 81.11% ±0.64 69.26% ±0.98 11.85 pts Reported ±1 SE ranges do not overlap
MysteryMechanism 49.10% ±3.36 21.62% ±2.77 27.48 pts Reported ±1 SE ranges do not overlap
Terminal-Bench Science 38.57% ±5.86 5.71% ±2.79 32.86 pts Reported ±1 SE ranges do not overlap
SAGE 51.77% ±3.41 43.53% ±3.43 8.24 pts Reported ±1 SE ranges do not overlap
Code Migration 69.83% ±4.26 40.93% ±4.35 28.90 pts Reported ±1 SE ranges do not overlap
IOI 83.06% ±3.74 47.67% ±4.79 35.39 pts Reported ±1 SE ranges do not overlap
Terminal-Bench 4.0 53.03% ±1.51 21.21% ±1.75 31.82 pts Reported ±1 SE ranges do not overlap
Vibe Code Bench v1.1 92.39% ±1.26 78.96% ±4.04 13.43 pts Reported ±1 SE ranges do not overlap
CyberBench v1.1 59.58% ±5.21 75.36% ±5.42 15.77 pts Reported ±1 SE ranges do not overlap
SRE Bench 30.15% ±2.84 4.20% ±1.24 25.95 pts Reported ±1 SE ranges do not overlap
Public Benefits Bench v1.1 67.19% ±1.22 67.59% ±1.22 0.41 pts Reported ±1 SE ranges overlap

Performance by category

Category Claude Sonnet 5.5 average MiMo V2.6 Flash average
Index 69.22% 59.58%
Legal 25.50% 24.62%
Finance 69.07% 60.55%
Healthcare 72.01% 63.17%
Math 100.00% 63.00%
Science 56.26% 32.20%
Education 51.77% 43.53%
Coding 74.58% 47.19%
Beta 44.87% 39.78%
Social Mobility 67.19% 67.59%

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 Claude Sonnet 5.5 cost MiMo V2.6 Flash cost Claude Sonnet 5.5 latency MiMo V2.6 Flash latency
Vals Index $20.80 $0.20 1h10m 45m02s
Harvey's Legal Agent Benchmark $16.61 $0.09 57m44s 17m12s
Legal Research Bench $14.83 $0.07 1h33m 15m20s
EMB $8.79 $0.11 36m45s 25m22s
Finance Agent (v2) $6.55 $0.07 31m43s 7m02s
Tax Agent Bench $9.25 $0.05 59m21s 17m19s
MedCode N/A N/A 3m59s 93.37s
MedScribe N/A N/A 5m07s 74.48s
ProofBench v1.1 $0.56 $0.12 5m48s 1h03m
BioMysteryBench $3.74 $0.05 18m06s 28m28s
MysteryMechanism $3.45 $0.06 23m30s 35m08s
Terminal-Bench Science $31.98 $0.31 3h22m 3h37m
SAGE N/A N/A 113.42s 2m16s
Code Migration $75.83 $0.49 3h20m 3h12m
IOI $7.52 $0.22 45m51s 1h49m
Terminal-Bench 4.0 $19.33 $0.22 2h06m 2h21m
Vibe Code Bench v1.1 $31.25 $0.56 1h01m 50m13s
CyberBench v1.1 $4.42 $0.05 33m53s 28m22s
SRE Bench $26.52 $0.40 1h55m 3h22m
Public Benefits Bench v1.1 $4.25 $0.03 1h18m 23m17s

Results available only for Claude Sonnet 5.5

None.

Results available only for MiMo V2.6 Flash

  • ProgramBench
Model details Claude Sonnet 5.5 Model details MiMo V2.6 Flash