Skip to content
fedi.software

Aya Expanse 32B in fedi-index

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

fedi-index · General

D 1,216 ±3

Percentile 24 · #226 · provisional

People prefer / Solves tasks

1,217 / —

Human votes vs benchmarks, Elo-eq

Value

—

On your hardware

Closed weights: cloud only.

Score decomposition

D General 1,216 [1,213–1,219] Percentile 24 · #226 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,224 ±2.48 1,224 ±2 26.1% 2026-11-11 —
LMArena text/hard_prompts 1,193 ±3.83 1,208 ±4 25.4% 2026-11-11 —
LMArena text/instruction_following 1,196 ±3.40 1,220 ±3 12.8% 2026-11-11 —
LMArena text/expert 1,182 ±7.08 1,216 ±6 11.6% 2026-11-11 —
LMArena text/multi_turn 1,190 ±4.64 1,208 ±4 6.2% 2026-11-11 —
LMArena text/longer_query 1,227 ±4.77 1,239 ±4 6.2% 2026-11-11 —

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

D Coding 1,200 [1,193–1,207] Percentile 21 · #227 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,196 ±4.63 1,200 ±4 88.1% 2026-11-11 —

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

D Math 1,201 [1,193–1,209] Percentile 23 · #197 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,200 ±4.93 1,201 ±5 87.9% 2026-11-11 —

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

C Writing 1,227 [1,220–1,234] Percentile 27 · #194 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,200 ±5.07 1,224 ±5 61.8% 2026-11-11 —
LMArena text/longer_query 1,227 ±4.77 1,239 ±4 31.8% 2026-11-11 —

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

D Language: 繁體中文 1,222 [1,213–1,232] Percentile 18 · #180 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,211 ±6.37 1,224 ±5 93.5% 2026-11-11 —

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

D Language: Deutsch 1,226 [1,207–1,245] Percentile 14 · #125 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,198 ±12 1,228 ±11 91.4% 2026-11-11 —

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

C Language: English 1,207 [1,201–1,213] Percentile 26 · #198 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,229 ±3.12 1,207 ±3 93.8% 2026-11-11 —

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

D Language: 日本語 1,246 [1,230–1,262] Percentile 11 · #113 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,163 ±12 1,250 ±9 92.3% 2026-11-11 —

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

D Language: 한국어 1,247 [1,225–1,268] Percentile 10 · #123 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,158 ±16 1,252 ±12 90.6% 2026-11-11 —

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

D Language: Русский 1,248 [1,240–1,256] Percentile 20 · #185 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,227 ±4.99 1,251 ±4 93.6% 2026-11-11 —

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 2026-11
Aya Expanse 32B 1,216 D

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: Aya Expanse 32B

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