Claude Fable 5 vs MiMo V2.6 Flash: Benchmark Comparison
Claude Fable 5 has the higher score on 15 of 16 shared benchmarks; MiMo V2.6 Flash leads on 0.
The largest observed score gap is 32.00 pts on ProofBench v1.1 , where Claude Fable 5 leads.
Reported ±1 standard-error ranges overlap on 3 of 16 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 Fable 5 | MiMo V2.6 Flash | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 61.39% ±1.00 | 53.23% ±1.10 | 8.16 pts | Reported ±1 SE ranges do not overlap |
| Harvey's Legal Agent Benchmark | 11.25% ±2.15 | 11.25% ±2.54 | 0.00 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 49.52% ±3.48 | 37.98% ±3.37 | 11.54 pts | Reported ±1 SE ranges do not overlap |
| EMB | 73.67% ±2.47 | 65.46% ±2.71 | 8.21 pts | Reported ±1 SE ranges do not overlap |
| Finance Agent (v2) | 56.31% ±0.84 | 56.28% ±0.46 | 0.04 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 69.82% ±3.12 | 59.90% ±3.13 | 9.92 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 56.07% ±2.20 | 41.06% ±2.03 | 15.01 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 88.52% ±1.95 | 85.28% ±1.99 | 3.25 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 95.00% ±2.19 | 63.00% ±4.85 | 32.00 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench Science | 15.71% ±4.38 | 4.29% ±2.44 | 11.43 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 51.89% ±3.40 | 43.53% ±3.43 | 8.36 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 55.06% ±4.61 | 40.93% ±4.35 | 14.14 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 2.00% ±0.99 | 0.50% ±0.50 | 1.50 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 41.41% ±1.34 | 24.24% ±1.51 | 17.17 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 90.35% ±2.10 | 78.96% ±4.04 | 11.39 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 70.43% ±1.19 | 67.59% ±1.22 | 2.84 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Fable 5 average | MiMo V2.6 Flash average |
|---|---|---|
| Legal | 30.38% | 24.62% |
| Finance | 66.60% | 60.55% |
| Healthcare | 72.30% | 63.17% |
| Math | 95.00% | 63.00% |
| Science | 15.71% | 4.29% |
| Education | 51.89% | 43.53% |
| Coding | 47.21% | 36.16% |
| Social Mobility | 70.43% | 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 Fable 5 cost | MiMo V2.6 Flash cost | Claude Fable 5 latency | MiMo V2.6 Flash latency |
|---|---|---|---|---|
| Vals Index | $29.59 | $0.20 | 42m24s | 58m02s |
| Harvey's Legal Agent Benchmark | $19.23 | $0.09 | 26m53s | 17m12s |
| Legal Research Bench | $9.79 | $0.07 | 22m18s | 15m20s |
| EMB | $12.35 | $0.11 | 26m34s | 25m22s |
| Finance Agent (v2) | $8.06 | $0.07 | 10m11s | 7m02s |
| Tax Agent Bench | $6.70 | $0.05 | 14m27s | 17m19s |
| MedCode | N/A | N/A | 91.44s | 93.37s |
| MedScribe | N/A | N/A | 119.47s | 74.48s |
| ProofBench v1.1 | $5.45 | $0.12 | 13m43s | 1h03m |
| Terminal-Bench Science | $51.99 | $0.24 | 1h59m | 3h59m |
| SAGE | N/A | N/A | 116.92s | 2m16s |
| Code Migration | $112.10 | $0.49 | 1h49m | 3h12m |
| ProgramBench | $75.68 | $0.47 | 2h37m | 3h22m |
| Terminal-Bench 4.0 | $30.34 | $0.22 | 1h08m | 2h25m |
| Vibe Code Bench v1.1 | $41.71 | $0.56 | 1h02m | 50m13s |
| Public Benefits Bench v1.1 | $4.59 | $0.03 | 21m17s | 20m56s |
Results available only for Claude Fable 5
- Vals RSI Index
- LegalBench
- MortgageTax
- TaxEval v2
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- LiveCodeBench
- SWE-bench
Results available only for MiMo V2.6 Flash
- BioMysteryBench
- MysteryMechanism
- IOI
- CyberBench v1.1
- SRE Bench