Claude Fable 5.1 vs GPT-6.1 Sol: Benchmark Comparison
Claude Fable 5.1 has the higher score on 15 of 19 shared benchmarks; GPT-6.1 Sol leads on 4.
The largest observed score gap is 31.13 pts on CyberBench v1.1 , where Claude Fable 5.1 leads.
Reported ±1 standard-error ranges overlap on 6 of 17 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 5.1 | GPT-6.1 Sol | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals RSI Index | 36.09% | 26.62% | 9.47 pts | Uncertainty comparison unavailable |
| Legal Research Bench | 55.29% ±3.46 | 38.46% ±3.38 | 16.83 pts | Reported ±1 SE ranges do not overlap |
| EMB v1.1 | 76.73% ±2.13 | 71.09% ±2.75 | 5.64 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 49.19% ±3.61 | 21.75% ±2.98 | 27.44 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 53.51% ±2.17 | 48.84% ±2.12 | 4.67 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 91.29% ±1.95 | 86.45% ±1.90 | 4.84 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 100.00% ±0.00 | 99.00% ±1.00 | 1.00 pts | Reported ±1 SE ranges overlap |
| MysteryMechanism | 47.75% ±3.36 | 46.40% ±3.35 | 1.35 pts | Reported ±1 SE ranges overlap |
| Robo-bench | 5.83% | 28.33% | 22.50 pts | Uncertainty comparison unavailable |
| Terminal-Bench-Science | 40.00% ±5.90 | 52.86% ±6.01 | 12.86 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 48.53% ±3.34 | 46.54% ±3.46 | 1.99 pts | Reported ±1 SE ranges overlap |
| IOI | 90.78% ±4.65 | 96.89% ±3.11 | 6.11 pts | Reported ±1 SE ranges overlap |
| ProgramBench | 7.00% ±1.81 | 3.00% ±1.21 | 4.00 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench Multimodal | 61.25% ±2.23 | 43.54% ±2.27 | 17.71 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 58.08% ±3.31 | 55.05% ±1.82 | 3.03 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 90.26% ±1.57 | 88.93% ±1.89 | 1.33 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 70.42% ±5.43 | 39.29% ±2.74 | 31.13 pts | Reported ±1 SE ranges do not overlap |
| SRE Bench | 22.90% ±2.60 | 50.76% ±3.10 | 27.86 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 74.90% ±1.13 | 59.34% ±1.28 | 15.56 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Fable 5.1 average | GPT-6.1 Sol average |
|---|---|---|
| Legal | 55.29% | 38.46% |
| Finance | 62.96% | 46.42% |
| Healthcare | 72.40% | 67.65% |
| Math | 100.00% | 99.00% |
| Science | 31.19% | 42.53% |
| Education | 48.53% | 46.54% |
| Coding | 61.47% | 57.48% |
| Cyber | 46.66% | 45.02% |
| Social Mobility | 74.90% | 59.34% |
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 5.1 cost | GPT-6.1 Sol cost | Claude Fable 5.1 latency | GPT-6.1 Sol latency |
|---|---|---|---|---|
| Vals RSI Index | $298.22 | $98.21 | 90h00m | 90h00m |
| Legal Research Bench | $23.06 | $3.38 | 57m29s | 34m20s |
| EMB v1.1 | $15.93 | $1.14 | 29m32s | 18m28s |
| Tax Agent Bench | $13.18 | $1.32 | 28m34s | 20m14s |
| MedCode | N/A | N/A | 3m35s | 115.34s |
| MedScribe | N/A | N/A | 3m08s | 2m50s |
| ProofBench v1.1 | $2.70 | $0.23 | 7m58s | 4m45s |
| MysteryMechanism | $5.63 | $0.28 | 17m41s | 7m17s |
| Robo-bench | $20.40 | $4.97 | 2m04s | 79.89s |
| Terminal-Bench-Science | $38.01 | $3.44 | 2h36m | 1h23m |
| SAGE | N/A | N/A | 105.75s | 50.22s |
| IOI | $11.20 | $1.26 | 30m06s | 31m31s |
| ProgramBench | $58.52 | $2.87 | 2h27m | 1h10m |
| SWE-bench Multimodal | $5.46 | $0.35 | 15m12s | 6m56s |
| Terminal-Bench 4.0 | $17.18 | $1.72 | 1h00m | 45m12s |
| Vibe Code Bench v1.1 | $33.37 | $6.23 | 57m40s | 1h15m |
| CyberBench v1.1 | $4.01 | $0.29 | 18m32s | 3m42s |
| SRE Bench | $32.23 | $2.69 | 2h03m | 32m32s |
| Public Benefits Bench v1.1 | $7.48 | $0.69 | 48m02s | 50m24s |
Results available only for Claude Fable 5.1
- LegalBench
- MortgageTax
- TaxEval v2
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- LiveCodeBench
- SkillsBench
- Vibe Code Bench 1-100
- CUA-bench
- Time Horizon Index: KSP
Results available only for GPT-6.1 Sol
- Vals Finance Index
- Vals Index
- Vals Legal Index
- Vals Science Index
- HLAB-Vals-1.1
- ClaimBench
- Finance Agent v2.1
- BioMysteryBench
- Code Migration v1.1