Claude Opus 5.5 vs MiMo V2.6 Flash: Benchmark Comparison
Claude Opus 5.5 has the higher score on 19 of 21 shared benchmarks; MiMo V2.6 Flash leads on 2.
The largest observed score gap is 47.39 pts on IOI , where Claude Opus 5.5 leads.
Reported ±1 standard-error ranges overlap on 1 of 21 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 Opus 5.5 | MiMo V2.6 Flash | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 66.97% ±0.89 | 53.23% ±1.10 | 13.74 pts | Reported ±1 SE ranges do not overlap |
| Harvey's Legal Agent Benchmark | 3.75% ±1.47 | 11.25% ±2.54 | 7.50 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 50.48% ±3.48 | 37.98% ±3.37 | 12.50 pts | Reported ±1 SE ranges do not overlap |
| EMB | 75.94% ±2.38 | 65.46% ±2.71 | 10.48 pts | Reported ±1 SE ranges do not overlap |
| Finance Agent (v2) | 58.59% ±0.17 | 56.28% ±0.46 | 2.31 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 70.50% ±3.15 | 59.90% ±3.13 | 10.59 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 49.80% ±2.27 | 41.06% ±2.03 | 8.74 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 91.43% ±1.93 | 85.28% ±1.99 | 6.16 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 | 79.26% ±0.98 | 69.26% ±0.98 | 10.00 pts | Reported ±1 SE ranges do not overlap |
| MysteryMechanism | 49.55% ±3.36 | 21.62% ±2.77 | 27.93 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench Science | 47.14% ±6.01 | 4.29% ±2.44 | 42.86 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 45.83% ±3.36 | 43.53% ±3.43 | 2.30 pts | Reported ±1 SE ranges overlap |
| Code Migration | 66.65% ±4.33 | 40.93% ±4.35 | 25.72 pts | Reported ±1 SE ranges do not overlap |
| IOI | 95.06% ±4.94 | 47.67% ±4.79 | 47.39 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 18.50% ±2.75 | 0.50% ±0.50 | 18.00 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 65.15% ±0.00 | 24.24% ±1.51 | 40.91 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 90.29% ±1.53 | 78.96% ±4.04 | 11.33 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 55.36% ±5.13 | 75.36% ±5.42 | 20.00 pts | Reported ±1 SE ranges do not overlap |
| SRE Bench | 33.59% ±2.92 | 4.20% ±1.24 | 29.39 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 70.64% ±1.19 | 67.59% ±1.22 | 3.05 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Opus 5.5 average | MiMo V2.6 Flash average |
|---|---|---|
| Legal | 27.12% | 24.62% |
| Finance | 68.34% | 60.55% |
| Healthcare | 70.61% | 63.17% |
| Math | 100.00% | 63.00% |
| Science | 58.65% | 31.72% |
| Education | 45.83% | 43.53% |
| Coding | 67.13% | 38.46% |
| Cyber | 44.47% | 39.78% |
| Social Mobility | 70.64% | 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 Opus 5.5 cost | MiMo V2.6 Flash cost | Claude Opus 5.5 latency | MiMo V2.6 Flash latency |
|---|---|---|---|---|
| Vals Index | $32.14 | $0.20 | 1h19m | 58m02s |
| Harvey's Legal Agent Benchmark | $21.38 | $0.09 | 49m00s | 17m12s |
| Legal Research Bench | $25.16 | $0.07 | 1h52m | 15m20s |
| EMB | $11.02 | $0.11 | 35m06s | 25m22s |
| Finance Agent (v2) | $9.22 | $0.07 | 33m08s | 7m02s |
| Tax Agent Bench | $15.09 | $0.05 | 1h07m | 17m19s |
| MedCode | N/A | N/A | 4m07s | 93.37s |
| MedScribe | N/A | N/A | 7m22s | 74.48s |
| ProofBench v1.1 | $0.96 | $0.12 | 5m51s | 1h03m |
| BioMysteryBench | $3.64 | $0.05 | 12m19s | 28m28s |
| MysteryMechanism | $4.62 | $0.06 | 19m38s | 35m08s |
| Terminal-Bench Science | $19.12 | $0.24 | 2h28m | 3h59m |
| SAGE | N/A | N/A | 114.20s | 2m16s |
| Code Migration | $112.97 | $0.49 | 3h19m | 3h12m |
| IOI | $5.26 | $0.22 | 17m15s | 1h49m |
| ProgramBench | $68.05 | $0.47 | 2h24m | 3h22m |
| Terminal-Bench 4.0 | $13.20 | $0.22 | 1h04m | 2h25m |
| Vibe Code Bench v1.1 | $57.92 | $0.56 | 1h36m | 50m13s |
| CyberBench v1.1 | $3.48 | $0.05 | 18m11s | 28m22s |
| SRE Bench | $34.65 | $0.40 | 1h37m | 3h22m |
| Public Benefits Bench v1.1 | $5.95 | $0.03 | 1h20m | 20m56s |
Results available only for Claude Opus 5.5
- Vals RSI Index
- Vibe Code Bench 1-100
- CUA-bench
- Time Horizon Index: KSP
Results available only for MiMo V2.6 Flash
None.