Claude Fable 5 vs GPT-5.6 Sol: Benchmark Comparison
Claude Fable 5 has the higher score on 21 of 25 shared benchmarks; GPT-5.6 Sol leads on 4.
The largest observed score gap is 12.10 pts on MedCode , where Claude Fable 5 leads.
Reported ±1 standard-error ranges overlap on 12 of 24 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 | GPT-5.6 Sol | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 61.39% ±1.00 | 58.01% ±1.02 | 3.38 pts | Reported ±1 SE ranges do not overlap |
| Vals RSI Index | 24.32% | 23.88% | 0.44 pts | Uncertainty comparison unavailable |
| Harvey's Legal Agent Benchmark | 11.25% ±2.15 | 2.50% ±0.83 | 8.75 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 49.52% ±3.48 | 48.08% ±3.47 | 1.44 pts | Reported ±1 SE ranges overlap |
| LegalBench | 88.56% ±0.33 | 86.97% ±0.41 | 1.60 pts | Reported ±1 SE ranges do not overlap |
| EMB | 73.67% ±2.47 | 72.34% ±2.37 | 1.33 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 56.31% ±0.84 | 53.76% ±0.85 | 2.56 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 68.92% ±0.91 | 67.29% ±0.92 | 1.63 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 69.82% ±3.12 | 67.95% ±3.12 | 1.87 pts | Reported ±1 SE ranges overlap |
| TaxEval v2 | 76.94% ±0.82 | 74.78% ±0.86 | 2.17 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 56.07% ±2.20 | 43.97% ±2.26 | 12.10 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 88.52% ±1.95 | 85.23% ±1.97 | 3.29 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 95.00% ±2.19 | 83.00% ±3.77 | 12.00 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench Science | 15.71% ±4.38 | 20.00% ±4.82 | 4.29 pts | Reported ±1 SE ranges overlap |
| GPQA Diamond | 93.18% ±1.94 | 95.20% ±1.07 | 2.02 pts | Reported ±1 SE ranges overlap |
| MMLU Pro | 91.50% ±0.28 | 89.10% ±0.31 | 2.40 pts | Reported ±1 SE ranges do not overlap |
| MMMU Pro | 89.31% ±0.74 | 88.84% ±0.76 | 0.46 pts | Reported ±1 SE ranges overlap |
| SAGE | 51.89% ±3.40 | 52.56% ±3.42 | 0.67 pts | Reported ±1 SE ranges overlap |
| Code Migration | 55.06% ±4.61 | 52.92% ±4.35 | 2.15 pts | Reported ±1 SE ranges overlap |
| LiveCodeBench | 89.78% ±0.89 | 82.60% ±1.09 | 7.17 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 2.00% ±0.99 | 1.50% ±0.86 | 0.50 pts | Reported ±1 SE ranges overlap |
| SWE-bench | 95.00% ±0.98 | 96.20% ±0.86 | 1.20 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 41.41% ±1.34 | 37.88% ±0.00 | 3.54 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 90.35% ±2.10 | 80.50% ±3.72 | 9.86 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 70.43% ±1.19 | 66.51% ±1.23 | 3.92 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Fable 5 average | GPT-5.6 Sol average |
|---|---|---|
| Legal | 49.78% | 45.85% |
| Finance | 69.13% | 67.22% |
| Healthcare | 72.30% | 64.60% |
| Math | 95.00% | 83.00% |
| Science | 15.71% | 20.00% |
| Academic | 91.33% | 91.05% |
| Education | 51.89% | 52.56% |
| Coding | 62.27% | 58.60% |
| Social Mobility | 70.43% | 66.51% |
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 | GPT-5.6 Sol cost | Claude Fable 5 latency | GPT-5.6 Sol latency |
|---|---|---|---|---|
| Vals Index | $29.59 | $14.24 | 42m24s | 37m23s |
| Vals RSI Index | $491.22 | $347.10 | 90h00m | 90h00m |
| Harvey's Legal Agent Benchmark | $19.23 | $10.37 | 26m53s | 22m50s |
| Legal Research Bench | $9.79 | $21.61 | 22m18s | 1h17m |
| LegalBench | N/A | N/A | 8.96s | 6.20s |
| EMB | $12.35 | $6.01 | 26m34s | 9m50s |
| Finance Agent (v2) | $8.06 | $1.25 | 10m11s | 19m25s |
| MortgageTax | N/A | N/A | 16.10s | 12.62s |
| Tax Agent Bench | $6.70 | $7.51 | 14m27s | 44m41s |
| TaxEval v2 | N/A | N/A | 56.87s | 62.03s |
| MedCode | N/A | N/A | 91.44s | 96.59s |
| MedScribe | N/A | N/A | 119.47s | 94.40s |
| ProofBench v1.1 | $5.45 | $1.44 | 13m43s | 8m40s |
| Terminal-Bench Science | $51.99 | $7.32 | 1h59m | 1h44m |
| GPQA Diamond | N/A | N/A | 99.90s | 58.30s |
| MMLU Pro | N/A | N/A | 25.00s | 15.65s |
| MMMU Pro | N/A | N/A | 61.44s | 26.20s |
| SAGE | N/A | N/A | 116.92s | 75.39s |
| Code Migration | $112.10 | $24.54 | 1h49m | 59m12s |
| LiveCodeBench | N/A | N/A | 118.53s | 56.51s |
| ProgramBench | $75.68 | $15.29 | 2h37m | 29m24s |
| SWE-bench | $2.05 | $1.15 | 5m56s | 3m02s |
| Terminal-Bench 4.0 | $30.34 | $7.98 | 1h08m | 33m46s |
| Vibe Code Bench v1.1 | $41.71 | $33.40 | 1h02m | 33m53s |
| Public Benefits Bench v1.1 | $4.59 | $8.85 | 21m17s | 1h01m |
Results available only for Claude Fable 5
None.
Results available only for GPT-5.6 Sol
- BioMysteryBench
- MysteryMechanism
- IOI
- SkillsBench
- Vibe Code Bench 1-100
- CyberBench v1.1
- SRE Bench
- CUA-bench
- Time Horizon Index: KSP