Claude Fable 5.2 in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
A 1,525 ±19
Percentile 100 · #1 · provisional
People prefer / Solves tasks
— / 1,564
Human votes vs benchmarks, Elo-eq
Value
+66◆
$16 per 1M tokens, median of 23 providers · vendor $16
On your hardware
Closed weights: cloud only.
Score decomposition
A General 1,525 [1,506–1,544] Percentile 100 · #1 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
173 ±1.93 | 1,623 ±15 | 64.0% | 2026-11-09 | — |
| Epoch AI |
simpleqa_verified
decay ×0.58 |
0.94 ±0.01 | 1,593 ±2 | 12.2% | 2026-11-06 | max |
| Epoch AI |
gpqa_diamond
outlier decay ×0.25 |
0.94 ±0.02 | 1,476 ±8 | 2.2% | 2026-11-06 | 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 Coding 1,509 [1,505–1,513] Percentile 100 · #1 · components: 2
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
webdev/overall
|
1,826 ±9.59 | 1,542 ±3 | 67.2% | 2026-11-15 | xhigh |
| Terminal-Bench |
tbench/terminal-bench
decay ×0.90 |
0.66 ±0.01 | 1,525 ±3 | 24.2% | 2026-11-09 | 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 Agents 1,532 [1,522–1,544] Percentile 100 · #1 · components: 2
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
agent/overall
|
0.29 ±0.02 | 1,592 ±9 | 60.7% | 2026-11-15 | max |
| Terminal-Bench |
tbench/terminal-bench
decay ×0.90 |
0.66 ±0.01 | 1,525 ±3 | 28.9% | 2026-11-09 | 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 Math 1,491 [1,488–1,494] Percentile 98 · #5 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
frontiermath_t1_3
decay ×0.82 |
0.95 ±0.01 | 1,527 ±3 | 57.0% | 2026-11-06 | max |
| Epoch AI |
frontiermath_t4
decay ×0.82 |
1.00 ±0.00 | 1,552 ±0 | 29.5% | 2026-11-06 | 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
The same in a table
| Model | 2026-11 |
|---|---|
| Claude Fable 5.2 | 1,525 A |
Badge
Show the model's fedi-index on your site or in a README. The badge updates itself; it links to this page.
Data CC BY 4.0: keep the link to fedi.software.