Claude Fable 5 vs GPT-6 Luna
Claude Fable 5 
GPT-6 Luna Verdict
- Claude Fable 5 (thinking) leads on quality: 70.0 vs 61.2.
- Claude Fable 5 (thinking) is stronger in agents & tools, composite, instruction following, knowledge, reasoning.
- GPT-6 Luna (max) is stronger in long context, multimodal.
- GPT-6 Luna (max) is 100× cheaper ($0.200 vs $20.00 per 1M blended).
- GPT-6 Luna (max) streams 2.4× faster (157 vs 67 tokens per second).
| Metric | Claude Fable 5 (thinking) | GPT-6 Luna (max) |
|---|---|---|
| BenchLeader Index | 70.0 | 61.2 |
| Agents & tools score | 74.1 | – |
| Composite score | 88.8 | 73.9 |
| Instruction following score | 63.0 | – |
| Knowledge score | 82.9 | 63.2 |
| Long context score | 67.0 | 67.5 |
| Reasoning score | 89.0 | 76.1 |
| Multimodal score | – | 59.2 |
| Blended price $/M | $20.00 | $0.200 |
| Output speed | 67 tok/s | 157 tok/s |
| Time to first answer | 108.0 s | 142.6 s |
| Context window | 1M | 1.1M |
| AA Intelligence Index | 49.6 | 37.3 |
| IFBench | 63.5% | – |
| AA-LCR | 82.3% | 83.3% |
| MMMU-Pro | – | 75.5% |
| AA-Omniscience | 43.3 | 0.7 |
| Terminal-Bench Hard | 62.9% | – |
| GPQA Diamond (AA) | 92.6% | – |
| Humanity's Last Exam (AA) | 55.5% | 38.5% |
| SciCode (AA) | 61.0% | 54.6% |
| τ²-Bench Telecom (AA) | 98.5% | – |
| Kagi LLM Benchmark | 91.4% | – |
| CritPt | 28.6% | 19.4% |
| GDPval (AA) | 54.8% | 43.4% |
| τ²-Bench Banking (AA) | 38.1% | – |
| Analyst Agent (AA) | 48.8% | – |
| Terminal-Bench 4.0 (AA) | 42.4% | 12.6% |
| Terminal-Bench 2.1 (AA) | 84.6% | – |
| AutomationBench | 54.1% | 53.2% |
| GDP.pdf | 24.0% | 20.4% |
| MLCR | 64.4% | 16.1% |
| Harvey LAB | 93.6% | – |
| EnterpriseOps-Gym | 51.1% | – |
| AA-Omniscience: accuracy | 65.3% | 43.8% |
| AA-Omniscience: non-hallucination | 36.4% | 23.3% |
| AA-Briefcase | 1543 | 1299 |
Data as of 2026-09-23. Best configuration of each model; every score links to its source on the model pages.
Claude Fable 5 vs GPT-6 Luna: questions
- Is Claude Fable 5 better than GPT-6 Luna?
- Claude Fable 5 (thinking) leads on quality: 70.0 vs 61.2. The BenchLeader Index combines every independent quality benchmark; Claude Fable 5 (thinking) is ahead overall as of 2026-09-23, but check the category scores for your use.
- Which is cheaper, Claude Fable 5 or GPT-6 Luna?
- GPT-6 Luna is cheaper: $0.200 against $20.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Fable 5 or GPT-6 Luna?
- GPT-6 Luna streams faster: 157 against 67 output tokens per second.
- Which has the larger context window?
- GPT-6 Luna accepts more context: 1.1M against 1M tokens.