Skip to content
fedi.software

Aya Expanse 8B in fedi-index

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

fedi-index · General

D 1,186 ±4

Percentile 21 · #235 · provisional

People prefer / Solves tasks

1,184 / —

Human votes vs benchmarks, Elo-eq

Value

—

On your hardware

Closed weights: cloud only.

Score decomposition

D General 1,186 [1,182–1,190] Percentile 21 · #235 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,185 ±3.53 1,185 ±4 27.1% 2026-11-11 —
LMArena text/hard_prompts 1,156 ±5.73 1,174 ±5 25.5% 2026-11-11 —
LMArena text/instruction_following 1,155 ±4.85 1,181 ±5 13.1% 2026-11-11 —
LMArena text/expert 1,152 ±12 1,192 ±10 10.0% 2026-11-11 —
LMArena text/multi_turn 1,160 ±7.33 1,181 ±7 6.0% 2026-11-11 —
LMArena text/longer_query 1,191 ±7.51 1,205 ±7 5.9% 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

D Coding 1,172 [1,161–1,183] Percentile 17 · #238 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,160 ±7.36 1,168 ±7 87.0% 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 1,175 [1,162–1,188] Percentile 20 · #206 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,168 ±7.69 1,171 ±7 86.6% 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 Writing 1,197 [1,186–1,207] Percentile 22 · #206 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,168 ±7.62 1,191 ±8 61.0% 2026-11-11 —
LMArena text/longer_query 1,191 ±7.51 1,205 ±7 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 1,200 [1,183–1,217] Percentile 14 · #188 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,181 ±12 1,200 ±9 92.0% 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: English 1,169 [1,161–1,178] Percentile 22 · #210 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,191 ±4.44 1,167 ±5 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

D Language: Русский 1,223 [1,211–1,235] Percentile 15 · #195 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,197 ±7.50 1,224 ±7 93.0% 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 8B 1,186 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 8B

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