Model comparison
Claude Fable 5.1 vs Qwen3.5 27B
Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 41.9 on the Noometry Index. Qwen3.5 27B costs 24× less per token, which makes it the better buy when Claude Fable 5.1's lead doesn't matter for your workload.
Last verified . 25 shared benchmarks.
Alibaba (Qwen)Summary
- They share 25 benchmarks with published results for both. Claude Fable 5.1 scores higher in 9 categories and Qwen3.5 27B in 0 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Fable 5.1 leads 89.6 to 38.8.
- The biggest single-benchmark swing is WeirdML: 92.9% for Claude Fable 5.1 and 39.5% for Qwen3.5 27B.
- Qwen3.5 27B is cheaper at $0.30 / $2.40 per million input/output tokens, against $10 / $50 for Claude Fable 5.1.
- Claude Fable 5.1 accepts more context: 1M tokens versus 262K.
- Qwen3.5 27B has downloadable open weights; the other is API-only.
Side by side
| Claude Fable 5.1 | Qwen3.5 27B | |
|---|---|---|
| Provider | Anthropic | Alibaba (Qwen) |
| Noometry Index | 69.0 | 41.9 |
| Released | 2026-09-01 | 2026-02-23 |
| Weights | Proprietary | Open |
| Context window | 1M | 262K |
| Max output | 128K | 66K |
| Input $ / M tokens | $10 | $0.30 |
| Output $ / M tokens | $50 | $2.40 |
| Results tracked | 52 | 28 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Fable 5.1 leads
Claude Fable 5.1: 74.7 (#1), Qwen3.5 27B: 38.9 (#168)
| Benchmark | Claude Fable 5.1 | Qwen3.5 27B |
|---|---|---|
| LMArena WebDev | 1744 | 1358 |
| WeirdML | 92.9% | 39.5% |
| LMArena Coding | 1528 | 1427 |
| ALE-Bench | 2,143 | 349.45 |
| FrontierCode | 50.9% | — |
| CursorBench | 51.8% | — |
| FrontierSWE | 56.3% | — |
| SciCode | 63.1% | — |
| GSO | 88.2% | — |
| MirrorCode | 73.3% | — |
Agentic & Tool Use Not comparable
Claude Fable 5.1: 50.7 (#5), Qwen3.5 27B: —
| Benchmark | Claude Fable 5.1 | Qwen3.5 27B |
|---|---|---|
| Vending-Bench 2 | 5,422 | 201.98 |
| APEX-Agents | 68.6% | — |
| Remote Labor Index | 17.9% | — |
| GDP.pdf | 29.6% | — |
Reasoning Claude Fable 5.1 leads
Claude Fable 5.1: 76.7 (#7), Qwen3.5 27B: 27.5 (#117)
| Benchmark | Claude Fable 5.1 | Qwen3.5 27B |
|---|---|---|
| NYT Connections (extended) | 90% | 47.9% |
| LMArena Hard Prompts | 1526 | 1414 |
| DTBench | 97.6% | 82.4% |
| LMCA | 65.5% | 34% |
| ARC-AGI-2 | 90% | — |
| ARC-AGI-1 | 97.5% | — |
| CritPt | 31.1% | — |
| Chess Puzzles | 47% | — |
| Thematic Generalization | — | 45.5% |
| EBR-Bench | 57.1% | — |
| Mystery Game Puzzles | 58% | — |
| Epoch Capabilities Index | 164.7 | — |
Math Claude Fable 5.1 leads
Claude Fable 5.1: 89.6 (#4), Qwen3.5 27B: 38.8 (#127)
| Benchmark | Claude Fable 5.1 | Qwen3.5 27B |
|---|---|---|
| LMArena Math | 1525 | 1429 |
| FrontierMath (Tiers 1-3) | 90.2% | — |
| FrontierMath Tier 4 | 87.8% | — |
| MathArena Final-Answer Competitions | — | 56.7% |
| OTIS Mock AIME 2024-2025 | 100% | — |
| ProofBench | 100% | — |
| FrontierMath Erdős | 0% | — |
Knowledge Claude Fable 5.1 leads
Claude Fable 5.1: 69.6 (#6), Qwen3.5 27B: 38.0 (#150)
| Benchmark | Claude Fable 5.1 | Qwen3.5 27B |
|---|---|---|
| LMArena Expert | 1535 | 1428 |
| Humanity's Last Exam | 46.5% | — |
| SimpleQA Verified | 70.8% | — |
| Vectara Hallucination Rate | — | 12.1% |
Multimodal Claude Fable 5.1 leads
Claude Fable 5.1: 53.9 (#4), Qwen3.5 27B: 39.4 (#59)
| Benchmark | Claude Fable 5.1 | Qwen3.5 27B |
|---|---|---|
| LMArena Vision | 1318 | 1241 |
| Blueprint-Bench 2 | 41.9% | — |
| Furniture Assembly | 70% | — |
| LMArena Document | 1513 | — |
Multilingual Claude Fable 5.1 leads
Claude Fable 5.1: 59.1 (#3), Qwen3.5 27B: 50.8 (#115)
| Benchmark | Claude Fable 5.1 | Qwen3.5 27B |
|---|---|---|
| LMArena Non-English | 1507 | 1390 |
| LMArena Chinese | 1586 | 1478 |
| LMArena French | 1525 | 1410 |
| LMArena German | 1500 | 1393 |
| LMArena Japanese | 1543 | 1345 |
| LMArena Korean | 1534 | 1358 |
| LMArena Russian | 1521 | 1390 |
| LMArena Spanish | 1516 | 1407 |
Instruction Following Claude Fable 5.1 leads
Claude Fable 5.1: 79.2 (#6), Qwen3.5 27B: 73.5 (#119)
| Benchmark | Claude Fable 5.1 | Qwen3.5 27B |
|---|---|---|
| LMArena Instruction Following | 1517 | 1393 |
Long Context Claude Fable 5.1 leads
Claude Fable 5.1: 46.7 (#20), Qwen3.5 27B: 43.1 (#106)
| Benchmark | Claude Fable 5.1 | Qwen3.5 27B |
|---|---|---|
| LMArena Longer Query | 1522 | 1413 |
Writing & Preference Claude Fable 5.1 leads
Claude Fable 5.1: 79.2 (#2), Qwen3.5 27B: 59.3 (#111)
| Benchmark | Claude Fable 5.1 | Qwen3.5 27B |
|---|---|---|
| LMArena Text | 1510 | 1409 |
| LMArena Creative Writing | 1507 | 1362 |
| LMArena Multi-Turn | 1492 | 1410 |
| EQ-Bench Creative Writing | 2162 | — |
Frequently asked questions
Is Claude Fable 5.1 better than Qwen3.5 27B?
Claude Fable 5.1 is the stronger model overall, scoring 69.0 to 41.9 on the Noometry Index. Qwen3.5 27B costs 24× less per token, which makes it the better buy when Claude Fable 5.1's lead doesn't matter for your workload.
Which is cheaper, Claude Fable 5.1 or Qwen3.5 27B?
Qwen3.5 27B is cheaper. It lists at $0.30 per million input tokens and $2.40 per million output tokens; Claude Fable 5.1 lists at $10 and $50.
Is Claude Fable 5.1 or Qwen3.5 27B better for coding?
Claude Fable 5.1 scores higher on coding benchmarks: 74.7 versus 38.9 in the Noometry coding category.
Which has the bigger context window?
Claude Fable 5.1 does, with 1M tokens against 262K.
How many benchmarks do Claude Fable 5.1 and Qwen3.5 27B share?
25 benchmarks have published results for both models. Claude Fable 5.1 has 52 scored results on Noometry and Qwen3.5 27B has 28.