Compare / Claude Fable 5 vs GPT-5.6 Terra
Claude Fable 5 vs GPT-5.6 Terra
31 results from 7 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Claude Fable 5 | GPT-5.6 Terra |
|---|---|---|---|
| APEX-Agentsreported | Tasks passed% of tasks | 63.6% | 58.2% |
| Arena (formerly LMArena)reported | Confirmed task successIPS effect estimate | 0.03 | -0.06 |
| Arena (formerly LMArena)reported | Praise over complaintIPS effect estimate | 0.17 | 0.03 |
| Arena (formerly LMArena)reported | SteerabilityIPS effect estimate | 0.11 | 0.04 |
| Arena (formerly LMArena)reported | Tool groundingIPS effect estimate | 0.00 | 0.00 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,499 | 1,477 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,483 | 1,467 |
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,503 | 1,466 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,501 | 1,428 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,550 | 1,504 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,509 | 1,462 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,504 | 1,465 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,508 | 1,445 |
| Artificial Analysisreported | Artificial Analysis Coding Indexindex score | 76.5 | 76.7 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 49.6 | 42.1 |
| Artificial Analysisreported | GPQA Diamond% of questions | 92.6% | 92.5% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 55.5% | 42.9% |
| Artificial Analysisreported | IFBench% of instructions | 63.5% | 71.2% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 82.3% | 83.0% |
| Artificial Analysisreported | SciCode% of problems | 61.0% | 55.0% |
| Artificial Analysisreported | Terminal-Bench 2.1% of tasks | 84.6% | 88.0% |
| Artificial Analysisreported | Terminal-Bench 4.0% of tasks | 42.4% | 35.4% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 62.9% | 62.9% |
| Artificial Analysisreported | Τ-bench banking% of tasks | 38.1% | 40.2% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 98.5% | 86.3% |
| GDP.pdfreported | Professional document tasks% of rubric | 30.0% | 24.7% |
| SimpleQA Verified (Epoch AI)reported | Correct answers% of questions | 70.7% | 43.2% |
| UGI Leaderboardreported | Requested-length error% off the requested word count | 2.0% | 22.0% |
| UGI Leaderboardreported | Style adherencescore from 0 to 1 | 0.41 | 0.39 |
| UGI Leaderboardreported | Writing scorescore out of 100 | 74.2 | 60.4 |
| Vending-Bench 2reported | Money after a yearUS dollars | $5,680.26 | $7,343.21 |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.