Claude Fable 5 vs Qwen3.5 397B A17B
Claude Fable 5 
Qwen3.5 397B A17B Verdict
- Claude Fable 5 leads on quality: 70.5 vs 58.7.
- Claude Fable 5 is stronger in agents & tools, coding, composite, human preference, knowledge, long context, multimodal, reasoning.
- Qwen3.5 397B A17B is stronger in instruction following, maths.
- Qwen3.5 397B A17B is 52× cheaper ($0.387 vs $20.00 per 1M blended).
| Metric | Claude Fable 5 | Qwen3.5 397B A17B |
|---|---|---|
| BenchLeader Index | 70.5 | 58.7 |
| Agents & tools score | 74.4 | 69.4 |
| Coding score | 70.5 | 51.7 |
| Composite score | 91.8 | 53.1 |
| Human preference score | 71.6 | 63.6 |
| Instruction following score | 62.9 | 76.1 |
| Knowledge score | 71.4 | 49.5 |
| Long context score | 67.8 | 65.2 |
| Multimodal score | 69.7 | 61.6 |
| Reasoning score | 75.9 | 63.0 |
| Maths score | – | 50.1 |
| Blended price $/M | $20.00 | $0.387 |
| Output speed | 63 tok/s | 78 tok/s |
| Time to first answer | 102.3 s | 42.7 s |
| Context window | 1M | 262k |
| GPQA Diamond | – | 85.9% |
| FrontierMath Tiers 1–3 | – | 29.5% |
| OTIS Mock AIME | – | 88.9% |
| Remote Labor Index | 16.1% | – |
| APEX-Agents | 45.0% | – |
| FrontierCode | 53.5% | – |
| Epoch Capabilities Index | 163.4 | 147.0 |
| LMArena Text | 1507 | 1441 |
| LMArena Hard Prompts | 1533 | 1463 |
| LMArena Coding | 1551 | 1491 |
| LMArena WebDev | 1628 | 1399 |
| LMArena Vision | 1330 | 1265 |
| AA Intelligence Index | 49.7 | 19.1 |
| IFBench | 63.5% | 78.8% |
| AA-LCR | 82.3% | 77.3% |
| MMMU-Pro | – | 77.3% |
| AA-Omniscience | 43.3 | -30.8 |
| Terminal-Bench Hard | 62.9% | 40.9% |
| GPQA Diamond (AA) | 92.6% | 89.3% |
| Humanity's Last Exam (AA) | 55.5% | 29.0% |
| SciCode (AA) | 61.0% | 44.8% |
| τ²-Bench Telecom (AA) | 98.5% | 95.6% |
| LiveCodeBench | 89.8% | – |
| MMLU-Pro | 91.5% | – |
| IOI | 72.3% | – |
| LegalBench | 88.6% | – |
| CorpFin | 71.8% | – |
| TaxEval | 76.9% | – |
| Terminal-Bench 2.1 (Vals) | 80.5% | – |
| SWE-bench (Vals) | 95.0% | – |
| GPQA Diamond (Vals) | 93.2% | – |
| Vals Index | 66.0 | – |
| AIME 2026 | – | 94.2% |
| HMMT February 2026 | – | 87.9% |
| MCP Atlas | 83.3% | – |
| PRBench Finance | 53.9% | – |
| PRBench Legal | 52.6% | – |
| HiL-Bench | 56.3% | – |
| EQ-Bench 4 | 1340 | – |
| Kagi LLM Benchmark | 88.8% | – |
Data as of 2026-09-10. Best configuration of each model; every score links to its source on the model pages.
Claude Fable 5 vs Qwen3.5 397B A17B: questions
- Is Claude Fable 5 better than Qwen3.5 397B A17B?
- Claude Fable 5 leads on quality: 70.5 vs 58.7. The BenchLeader Index combines every independent quality benchmark; Claude Fable 5 is ahead overall as of 2026-09-10, but check the category scores for your use.
- Is Claude Fable 5 better than Qwen3.5 397B A17B for coding?
- Claude Fable 5 scores higher in coding (71 vs 52 on the category index, where 50 is average).
- Is Claude Fable 5 better than Qwen3.5 397B A17B for agentic tasks?
- Claude Fable 5 scores higher in agentic tasks (74 vs 69 on the category index, where 50 is average).
- Which is cheaper, Claude Fable 5 or Qwen3.5 397B A17B?
- Qwen3.5 397B A17B is cheaper: $0.387 against $20.00 per million tokens, blended at three input tokens per output token.
- Which is faster, Claude Fable 5 or Qwen3.5 397B A17B?
- Qwen3.5 397B A17B streams faster: 78 against 63 output tokens per second.
- Which has the larger context window?
- Claude Fable 5 accepts more context: 1M against 262k tokens.