Skip to content
fedi.software

athene-70b-0725 in fedi-index

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

fedi-index · General

C 1,250 ±3

Percentile 28 · #211 · provisional

People prefer / Solves tasks

1,253 / —

Human votes vs benchmarks, Elo-eq

Value

—

On your hardware

Closed weights: cloud only.

Score decomposition

C General 1,250 [1,247–1,254] Percentile 28 · #211 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,265 ±2.88 1,265 ±3 26.5% 2026-10-02 —
LMArena text/hard_prompts 1,243 ±4.37 1,253 ±4 25.7% 2026-10-02 —
LMArena text/instruction_following 1,227 ±3.92 1,249 ±4 13.0% 2026-10-02 —
LMArena text/expert 1,228 ±9.82 1,254 ±8 10.7% 2026-10-02 —
LMArena text/multi_turn 1,257 ±5.53 1,269 ±5 6.2% 2026-10-02 —
LMArena text/longer_query 1,232 ±6.29 1,244 ±6 6.0% 2026-10-02 —

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 Coding 1,256 [1,248–1,265] Percentile 27 · #202 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,267 ±5.50 1,263 ±5 89.7% 2026-10-02 —

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 Math 1,228 [1,219–1,237] Percentile 27 · #186 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,231 ±5.33 1,231 ±5 87.8% 2026-10-02 —

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 1,253 [1,245–1,261] Percentile 30 · #188 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,238 ±5.74 1,263 ±6 62.2% 2026-10-02 —
LMArena text/longer_query 1,232 ±6.29 1,244 ±6 31.1% 2026-10-02 —

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 1,242 [1,231–1,253] Percentile 22 · #174 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,237 ±7.47 1,245 ±6 93.3% 2026-10-02 —

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 1,268 [1,249–1,287] Percentile 21 · #114 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,252 ±12 1,274 ±10 91.5% 2026-10-02 —

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 1,266 [1,259–1,272] Percentile 31 · #189 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,288 ±3.54 1,270 ±4 93.8% 2026-10-02 —

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 1,229 [1,213–1,245] Percentile 12 · #113 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,138 ±12 1,231 ±9 92.2% 2026-10-02 —

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 1,242 [1,220–1,263] Percentile 13 · #118 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,150 ±16 1,246 ±12 90.5% 2026-10-02 —

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 1,268 [1,257–1,279] Percentile 24 · #179 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,250 ±6.73 1,272 ±6 93.2% 2026-10-02 —

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
athene-70b-0725 1,250 C

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: athene-70b-0725

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