Claude Fable 5.1 vs GPT-5.6 Terra: Benchmark Comparison
Claude Fable 5.1 has the higher score on 25 of 27 shared benchmarks; GPT-5.6 Terra leads on 2.
The largest observed score gap is 35.35 pts on Terminal-Bench 4.0 , where Claude Fable 5.1 leads.
Reported ±1 standard-error ranges overlap on 7 of 27 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-5.6 Terra | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 65.83% ±1.11 | 53.09% ±1.29 | 12.74 pts | Reported ±1 SE ranges do not overlap |
| Harvey's Legal Agent Benchmark | 6.67% ±1.17 | 0.83% ±0.83 | 5.83 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 55.29% ±3.46 | 41.35% ±3.42 | 13.94 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 88.51% ±0.42 | 85.11% ±0.45 | 3.41 pts | Reported ±1 SE ranges do not overlap |
| EMB | 76.67% ±2.08 | 66.20% ±3.11 | 10.47 pts | Reported ±1 SE ranges do not overlap |
| Finance Agent (v2) | 58.88% ±2.06 | 54.44% ±2.07 | 4.44 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 70.79% ±0.90 | 67.33% ±0.93 | 3.46 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 77.64% ±2.83 | 65.20% ±3.19 | 12.44 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 75.96% ±0.83 | 76.17% ±0.84 | 0.20 pts | Reported ±1 SE ranges overlap |
| MedCode | 53.51% ±2.17 | 43.41% ±2.17 | 10.09 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 91.29% ±1.95 | 82.87% ±1.95 | 8.42 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 100.00% ±0.00 | 74.00% ±4.41 | 26.00 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench Science | 40.00% ±5.90 | 10.00% ±3.61 | 30.00 pts | Reported ±1 SE ranges do not overlap |
| GPQA Diamond | 93.43% ±1.86 | 90.91% ±1.91 | 2.52 pts | Reported ±1 SE ranges overlap |
| MMLU Pro | 92.38% ±0.27 | 86.66% ±0.33 | 5.72 pts | Reported ±1 SE ranges do not overlap |
| MMMU Pro | 90.64% ±0.70 | 86.47% ±0.82 | 4.16 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 48.53% ±3.34 | 47.00% ±3.40 | 1.52 pts | Reported ±1 SE ranges overlap |
| Code Migration | 54.61% ±4.81 | 47.80% ±4.28 | 6.80 pts | Reported ±1 SE ranges overlap |
| IOI | 90.78% ±4.65 | 87.61% ±6.53 | 3.17 pts | Reported ±1 SE ranges overlap |
| LiveCodeBench | 90.52% ±0.86 | 85.93% ±1.02 | 4.58 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 7.00% ±1.81 | 0.50% ±0.50 | 6.50 pts | Reported ±1 SE ranges do not overlap |
| SkillsBench | 61.55% ±4.76 | 58.90% ±4.47 | 2.66 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 58.08% ±3.31 | 22.73% ±2.31 | 35.35 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench 1-100 | 28.00% ±4.49 | 14.82% ±3.12 | 13.18 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 90.26% ±1.57 | 74.59% ±4.20 | 15.67 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 70.42% ±5.43 | 72.08% ±5.41 | 1.67 pts | Reported ±1 SE ranges overlap |
| Public Benefits Bench v1.1 | 74.90% ±1.13 | 62.38% ±1.26 | 12.52 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Fable 5.1 average | GPT-5.6 Terra average |
|---|---|---|
| Legal | 50.16% | 42.43% |
| Finance | 71.99% | 65.87% |
| Healthcare | 72.40% | 63.14% |
| Math | 100.00% | 74.00% |
| Science | 40.00% | 10.00% |
| Academic | 92.15% | 88.01% |
| Education | 48.53% | 47.00% |
| Coding | 60.10% | 49.11% |
| Cyber | 70.42% | 72.08% |
| Social Mobility | 74.90% | 62.38% |
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-5.6 Terra cost | Claude Fable 5.1 latency | GPT-5.6 Terra latency |
|---|---|---|---|---|
| Vals Index | $28.71 | $5.78 | 1h17m | 35m39s |
| Harvey's Legal Agent Benchmark | $46.21 | $3.20 | 1h52m | 17m05s |
| Legal Research Bench | $23.06 | $7.91 | 57m29s | 1h10m |
| LegalBench | N/A | N/A | 12.40s | 2.78s |
| EMB | $15.93 | $2.19 | 29m32s | 13m33s |
| Finance Agent (v2) | $8.35 | $3.63 | 18m14s | 24m10s |
| MortgageTax | N/A | N/A | 14.98s | 6.32s |
| Tax Agent Bench | $13.18 | $4.83 | 28m34s | 1h01m |
| TaxEval v2 | N/A | N/A | 85.56s | 37.28s |
| MedCode | N/A | N/A | 3m35s | 18.41s |
| MedScribe | N/A | N/A | 3m08s | 35.53s |
| ProofBench v1.1 | $2.70 | $1.27 | 7m58s | 12m57s |
| Terminal-Bench Science | $38.01 | $5.19 | 2h36m | 2h04m |
| GPQA Diamond | N/A | N/A | 45.86s | 36.76s |
| MMLU Pro | N/A | N/A | 20.59s | 8.17s |
| MMMU Pro | N/A | N/A | 37.47s | 13.03s |
| SAGE | N/A | N/A | 105.75s | 40.06s |
| Code Migration | $70.97 | $8.13 | 4h20m | 48m34s |
| IOI | $11.20 | $8.66 | 30m06s | 1h23m |
| LiveCodeBench | N/A | N/A | 59.74s | 44.33s |
| ProgramBench | $58.52 | $5.38 | 2h27m | 33m36s |
| SkillsBench | $4.52 | $1.75 | 11m02s | 6m55s |
| Terminal-Bench 4.0 | $17.18 | $5.60 | 1h00m | 31m29s |
| Vibe Code Bench 1-100 | $149.66 | $33.33 | 2h53m | 1h49m |
| Vibe Code Bench v1.1 | $33.37 | $7.89 | 57m40s | 19m12s |
| CyberBench v1.1 | $4.01 | $3.31 | 18m32s | 17m04s |
| Public Benefits Bench v1.1 | $7.48 | $1.20 | 48m02s | 19m31s |
Results available only for Claude Fable 5.1
- Vals RSI Index
- MysteryMechanism
- SRE Bench
- CUA-bench
- Time Horizon Index: KSP
Results available only for GPT-5.6 Terra
- SWE-bench