Claude Fable 5 vs Wonder Full: Benchmark Comparison
Claude Fable 5 has the higher score on 5 of 12 shared benchmarks; Wonder Full leads on 7.
The largest observed score gap is 21.43 pts on Terminal-Bench-Science , where Wonder Full leads.
Reported ±1 standard-error ranges overlap on 4 of 11 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 | Wonder Full | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals RSI Index | 24.32% | 24.43% | 0.11 pts | Uncertainty comparison unavailable |
| Legal Research Bench | 49.52% ±3.48 | 53.37% ±3.47 | 3.85 pts | Reported ±1 SE ranges overlap |
| LegalBench | 88.56% ±0.33 | 87.42% ±0.41 | 1.14 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 38.28% ±3.50 | 43.00% ±3.57 | 4.72 pts | Reported ±1 SE ranges overlap |
| MedCode | 56.07% ±2.20 | 49.93% ±2.18 | 6.14 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 88.52% ±1.95 | 89.33% ±2.00 | 0.81 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 95.00% ±2.19 | 94.00% ±2.39 | 1.00 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench-Science | 15.71% ±4.38 | 37.14% ±5.82 | 21.43 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 | 58.08% ±3.31 | 16.67 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 90.35% ±2.10 | 86.00% ±1.93 | 4.36 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 70.43% ±1.19 | 73.75% ±1.15 | 3.31 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Fable 5 average | Wonder Full average |
|---|---|---|
| Legal | 69.04% | 70.39% |
| Finance | 38.28% | 43.00% |
| Healthcare | 72.30% | 69.63% |
| Math | 95.00% | 94.00% |
| Science | 15.71% | 37.14% |
| Coding | 44.59% | 48.19% |
| Social Mobility | 70.43% | 73.75% |
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 | Wonder Full cost | Claude Fable 5 latency | Wonder Full latency |
|---|---|---|---|---|
| Vals RSI Index | N/A | N/A | 90h00m | 90h00m |
| Legal Research Bench | $9.79 | $3.60 | 22m18s | 27m34s |
| LegalBench | N/A | N/A | 8.96s | 38.27s |
| Tax Agent Bench | $6.70 | $2.15 | 14m27s | 18m04s |
| MedCode | N/A | N/A | 91.44s | 3m03s |
| MedScribe | N/A | N/A | 119.47s | 115.72s |
| ProofBench v1.1 | $5.45 | $0.86 | 13m43s | 11m36s |
| Terminal-Bench-Science | $51.99 | $22.81 | 1h59m | 3h35m |
| ProgramBench | $75.68 | $28.65 | 2h37m | 6h04m |
| Terminal-Bench 4.0 | $30.34 | $10.53 | 1h08m | 1h12m |
| Vibe Code Bench v1.1 | $41.71 | $2.29 | 1h02m | 10m48s |
| Public Benefits Bench v1.1 | $4.59 | $4.64 | 21m17s | 1h13m |
Results available only for Claude Fable 5
- MortgageTax
- TaxEval v2
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- SAGE
- LiveCodeBench
- SWE-bench
Results available only for Wonder Full
- Vals Finance Index
- Vals Index
- Vals Legal Index
- Vals Science Index
- HLAB-Vals-1.1
- EMB v1.1
- Finance Agent v2.1
- BioMysteryBench
- MysteryMechanism
- Robo-bench
- Code Migration v1.1
- IOI
- SWE-bench Multimodal
- Vibe Code Bench 1-100
- CyberBench v1.1
- SRE Bench