Claude Fable 5
AnthropicProprietaryReasoningVerifiedReleased Jun 9, 2026Input: text, image, file
Overall score
81.3
Rank 7 of 222 · 78.4–84.3
Context window
1M
1,000,000 tokens
Price per 1M tokens
$10.00 in · $50.00 out
✓ Verified on Anthropic's pricing page
Blended price
$20.00
3:1 input:output tokens
Output speed
—
Not measured yet
Category scores
Scores per capability area with the model's rank among ranked models in that category.
- 1Math96.0
- 2Coding#790.3
- 3Knowledge#685.0
- 4Reasoning84.3
- 5Agentic & tool use#682.9
- 6Instruction following#6277.1
- 7Multimodal66.0
Benchmark results (20)
As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.
Agentic & tool use
| Terminal-Bench 2.1 | 84.3 | #19 of 68 |
| MazeBench | 11 | #7 of 42 |
| OSWorld-Verified | 85.0 | #2 of 35 |
| ApprenticeBench | 34 | #4 of 18 |
| Terminal-Bench 3.0 | 34.0 | #3 of 13 |
Coding
| SWE-bench Verified | 95.0 | #3 of 98 |
| SWE-bench Pro | 80.0 | #5 of 69 |
| FrontierSWE v2 | 47.0 | #7 of 20 |
| FrontierCode 1.1 Main | 53.5 | #2 of 16 |
| VulcanBench v3 | 89.5 | #2 of 14 |
Multimodal
| Design Arena Website | 1297 | #16 of 91 |
| ParseBench | 70.78 | #9 of 37 |
| OfficeQA Pro | 57.9 | #9 of 13 |
| Blueprint-Bench 2 | 38.6 | #1 of 2 |
Reasoning
| ARC-AGI-2 | 89.2 | #9 of 41 |
| ARC-AGI-1 | 98.50 | #1 of 36 |
Math
| FrontierMath v2 (Tiers 1-3) | 87.018 | #9 of 64 |
| FrontierMath v2 (Tier 4) | 90.200 | #4 of 51 |
Pricing by source
USD per 1M tokens.
| Source | Input | Output | Cached in |
|---|---|---|---|
| ✓ Official (Anthropic) checked Oct 17, 2026 · Claude Fable 5 | $10.00 | $50.00 | $1.00 |
| BenchLM list price | $10.00 | $50.00 | $1.00 |
| OpenRouter anthropic/claude-fable-5 | $10.00 | $50.00 | $1.00 |
Anthropic's June 9, 2026 Claude Fable 5 model page says Claude Fable 5 is available through the Claude API as `claude-fable-5` at $10 input / $50 output per million tokens, with a 90% input-token discount for prompt caching. US-only inference is available at 1.1x pricing.
Arena ratings
Human-preference ratings from LMArena. Best-rated variant shown.
| Text(High) | 1491 | #10 · 42K votes |
| WebDev (coding)(High) | 1624 | #23 · 15.9K votes |
| Vision(High) | 1325 | #1 · 14.7K votes |
| Search(high) | 1229 | #10 · 73.6K votes |
| Documents | 1496 | #5 · 8.5K votes |
Speed
Measured by us through OpenRouter (median of trials).
Not measured yet.
Related models
Same family
Similar score
- Claude Sonnet 5.582.2 · $4.00
- Claude Opus 580.1 · $10.00
- Gemini 4 Argon82.9 · $4.00
- Claude Fable 5.183.2 · $20.00
- GPT-6 Astra84.0 · $20.00