Qwen3.6 27B is 26.1× cheaper per million tokens (blended). Claude Fable 5 has the larger context window (1,000,000 vs 262,144 tokens, 3.8×). Qwen3.6 27B is open-weights; the other is not.
Claude Fable 5 vs Qwen3.6 27B
Live catalog fields. Best-in-row is highlighted.
Claude Fable 5
Anthropic · anthropic__claude-fable-5
Qwen3.6 27Bopen
Qwen · qwen__qwen3.6-27b
Identity
| Field | anthropic__claude-fable-5 | qwen__qwen3.6-27b |
|---|---|---|
| Provider | Anthropic | Qwen |
| Slug | anthropic__claude-fable-5 | qwen__qwen3.6-27b |
| Status | live | live |
| Open weights | No | Yes |
| License | — | — |
| HuggingFace | — | Qwen/Qwen3.6-27B |
Quality
| Field | anthropic__claude-fable-5 | qwen__qwen3.6-27b |
|---|---|---|
| Quality index (0–100) | 89.4 | — |
Cost
| Field | anthropic__claude-fable-5 | qwen__qwen3.6-27b |
|---|---|---|
| Input $ / 1M tokens | $10.00 | $0.30 |
| Cached input $ / 1M | $1.00 | $0.03 |
| Output $ / 1M tokens | $50.00 | $2.00 |
Context
| Field | anthropic__claude-fable-5 | qwen__qwen3.6-27b |
|---|---|---|
| Max context tokens | 1,000,000 | 262,144 |
| Max output tokens | 128,000 | 65,536 |
Modalities
| Field | anthropic__claude-fable-5 | qwen__qwen3.6-27b |
|---|---|---|
| Input | text, image, file | text, image, video |
| Output | text | text |