Claude Fable 5
Anthropic Claude Fable 5 for long-running coding, research, and knowledge-work tasks. Anthropic retains prompts and outputs for 30 days under its Covered Models policy; Zero Data Retention is not available.
- Reasoning
- Vision
- Native PDF input
- Tool Calling
- Structured Output
Added Jun 9, 2026
Pricing
Auto routing · per 1M tokens- Input
- $10.00
- Output
- $50.00
- Cache read
- $1.00
Specifications
- Context window
- 1M
- Max output
- 128K
Benchmarks
Benchmarks
Sourced from LMArena.
Arena Score
1504.1
Overall Rank
#6 / 422
Votes
45,336
Confidence Interval
1499.9 - 1508.3
Category Scores
Coding
#4 / 417
11,607 votes
1551.3
Math
#4 / 405
2,238 votes
1520.2
Longer Query
#3 / 400
20,920 votes
1522.1
Creative Writing
#5 / 420
9,807 votes
1501.4
Instruction Following
#6 / 422
15,988 votes
1508.5
Hard Prompts
#7 / 422
29,440 votes
1529.5
Additional Categories22
Korean
#2 / 289
810 votes
1490.9
Expert
#3 / 372
4,866 votes
1550.2
French
#3 / 302
1,565 votes
1527.4
Industry Entertainment And Sports And Media
#4 / 420
11,827 votes
1495.0
Industry Life And Physical And Social Science
#5 / 420
7,303 votes
1525.3
Industry Writing And Literature And Language
#5 / 421
12,793 votes
1508.8
Japanese
#5 / 288
665 votes
1518.8
Multi Turn
#5 / 420
7,386 votes
1517.0
English
#6 / 422
19,169 votes
1512.8
Exclude Ties
#6 / 422
34,268 votes
1523.4
Hard Prompts English
#6 / 420
12,568 votes
1532.3
Industry Business And Management And Financial Operations
#6 / 415
8,936 votes
1500.6
Industry Software And It Services
#6 / 422
16,729 votes
1538.0
Polish
#6 / 243
697 votes
1509.3
Spanish
#6 / 304
1,433 votes
1494.6
Chinese
#7 / 396
3,322 votes
1554.8
Industry Legal And Government
#7 / 394
3,764 votes
1511.1
Non English
#7 / 422
26,164 votes
1492.7
Russian
#7 / 386
5,270 votes
1513.9
Industry Mathematical
#9 / 403
2,576 votes
1519.6
German
#16 / 321
725 votes
1496.9
Industry Medicine And Healthcare
#18 / 392
3,220 votes
1505.4
Published 2026-11-18 · Matched as Claude Fable 5 (High) · CC BY 4.0
LMArena DatasetLeaderboard results
| Benchmark | Score | Rank on NanoGPT |
|---|---|---|
Epoch AI Benchmarking Hub Epoch AI data, CC BY 4.0 · Updated Nov 19, 2026 | ||
Epoch Capabilities Index | 164.2 | #12 of 113 |
FrontierMath v2 (Tiers 1–3) | 87.0% | #13 of 83 |
GPQA Diamond | 85.9% | #66 of 118 |
SimpleQA Verified | 70.7% | #11 of 83 |
FrontierMath v2 (Tier 4) | 90.2% | #8 of 67 |
OTIS Mock AIME 2024–2025 | 100.0% | #1 of 120 |
MirrorCode | 70.0% | #7 of 41 |
Chess Puzzles | 47.0% | #23 of 82 |
Mystery Game Puzzles | 52.0% | #16 of 79 |
Sierra τ-bench tau2-bench data, MIT · Updated Aug 4, 2026 | ||
τ³-Banking | 39.7% | #7 of 18 |
EQ-Bench 4 | 1313 | #5 of 31 |
Creative Writing v3 | 1944 | #12 of 91 |
Longform Writing | 83.0 | #4 of 93 |
EQ-Bench 3 | 2096 | #1 of 45 |
UGI Leaderboard UGI Leaderboard data, Apache-2.0 · Updated Oct 16, 2026 | ||
UGI Score | 55.9 | #18 of 125 |
UGI Willingness | 1.8 | #106 of 125 |
UGI Writing | 74.2 | #4 of 125 |
Filtered and ranked to models on NanoGPT.
Providers
Auto routing is available for this model. Explicit provider selection is not available.
| Cache write /M | ||||||
|---|---|---|---|---|---|---|
Recommended | $10.00 | $50.00 | $1.00 | $12.50 | N/A | N/A |