Skip to content
fedi.software

Claude Fable 5.1 in fedi-index

The score in every slice, what it is made of, value, local run and history.

fedi-index · General

S 1,498 ±8

Percentile 100 · #2

People prefer / Solves tasks

1,490 / 1,512

Human votes vs benchmarks, Elo-eq

Value

+47

$20 per 1M tokens, median of 26 providers · vendor $20

On your hardware

Closed weights: cloud only.

Score decomposition

S General 1,498 [1,490–1,506] Percentile 100 · #2 · components: 8
Source Component Value Elo-eq Weight Measured Variant
Epoch AI eci 165 ±1.83 1,554 ±14 28.5% 2026-09-01 —
LMArena text/overall 1,511 ±3.26 1,511 ±3 17.5% 2026-10-02 max
LMArena text/hard_prompts 1,526 ±3.94 1,508 ±4 17.4% 2026-10-02 max
LMArena text/instruction_following 1,516 ±5.38 1,523 ±5 8.2% 2026-10-02 max
LMArena text/expert 1,534 ±9.65 1,504 ±8 7.2% 2026-10-02 max
Epoch AI simpleqa_verified decay ×0.61 0.71 ±0.01 1,520 ±4 5.1% 2026-09-01 max
LMArena text/longer_query 1,524 ±4.69 1,516 ±4 4.2% 2026-10-02 max
LMArena text/multi_turn 1,494 ±7.83 1,484 ±7 3.8% 2026-10-02 max

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

A Coding 1,477 [1,472–1,482] Percentile 99 · #3 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,749 ±4.88 1,520 ±1 47.8% 2026-10-01 max
LMArena text/coding 1,525 ±6.36 1,493 ±6 42.2% 2026-10-02 max

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

A Agents 1,491 [1,483–1,499] Percentile 100 · #1 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena agent/overall 0.14 ±0.01 1,524 ±5 89.8% 2026-10-02 max

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

S Math 1,489 [1,479–1,500] Percentile 100 · #2 · components: 3
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,521 ±13 1,509 ±12 40.5% 2026-10-02 max
Epoch AI frontiermath_t1_3 decay ×0.86 0.90 ±0.02 1,516 ±4 36.3% 2026-09-01 max
Epoch AI frontiermath_t4 decay ×0.86 0.88 ±0.05 1,533 ±8 14.8% 2026-09-01 max

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

A Vision 1,476 [1,469–1,482] Percentile 98 · #3 · components: 3 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena vision/overall 1,320 ±5.06 1,493 ±5 64.6% 2026-10-02 max
LMArena vision/ocr 1,328 ±5.72 1,498 ±6 15.6% 2026-10-02 max
LMArena vision/diagram 1,345 ±9.70 1,504 ±9 13.3% 2026-10-02 max

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

A Writing 1,507 [1,499–1,515] Percentile 99 · #4 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,505 ±6.39 1,536 ±6 60.3% 2026-10-02 max
LMArena text/longer_query 1,524 ±4.69 1,516 ±4 33.0% 2026-10-02 max

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

A Language: 繁體中文 1,499 [1,484–1,514] Percentile 100 · #2 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,588 ±11 1,523 ±8 92.4% 2026-10-02 max

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

A Language: English 1,478 [1,470–1,487] Percentile 98 · #5 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,504 ±4.53 1,497 ±5 93.6% 2026-10-02 max

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

A Language: Español 1,468 [1,443–1,492] Percentile 98 · #4 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,508 ±15 1,498 ±14 89.5% 2026-10-02 max

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

A Language: Français 1,468 [1,442–1,494] Percentile 98 · #4 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,524 ±16 1,500 ±15 89.0% 2026-10-02 max

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

A Language: Русский 1,496 [1,483–1,509] Percentile 100 · #2 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,524 ±8.02 1,518 ±7 92.9% 2026-10-02 max

Weight is the share of the component in the score (the rest is the prior that pulls models with little data towards the population median). Methodology

History

fedi-index by month
The same in a table
Model 2026-10
Claude Fable 5.1 1,498 S

Badge

Show the model's fedi-index on your site or in a README. The badge updates itself; it links to this page.

fedi-index: Claude Fable 5.1

Data CC BY 4.0: keep the link to fedi.software.