Skip to content
fedi.software

zephyr-7b-beta in fedi-index

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

fedi-index · General

D 1,082 ±5

Percentile 8 · #270 · provisional

People prefer / Solves tasks

1,074 / —

Human votes vs benchmarks, Elo-eq

Value

—

On your hardware

Closed weights: cloud only.

Score decomposition

D General 1,082 [1,077–1,088] Percentile 8 · #270 · components: 5 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,069 ±4.44 1,069 ±4 31.7% 2026-10-02 —
LMArena text/hard_prompts 1,027 ±7.25 1,058 ±6 29.0% 2026-10-02 —
LMArena text/instruction_following 1,032 ±6.46 1,064 ±6 14.8% 2026-10-02 —
LMArena text/multi_turn 1,038 ±8.93 1,070 ±8 6.7% 2026-10-02 —
LMArena text/longer_query 1,042 ±14 1,066 ±13 5.0% 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

D Coding 1,081 [1,067–1,095] Percentile 7 · #258 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,045 ±9.21 1,064 ±8 88.1% 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 Math 1,070 [1,056–1,084] Percentile 6 · #239 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,041 ±8.60 1,049 ±8 86.0% 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

D Writing 1,116 [1,103–1,128] Percentile 12 · #237 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,102 ±8.12 1,124 ±8 66.7% 2026-10-02 —
LMArena text/longer_query 1,042 ±14 1,066 ±13 25.2% 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: 繁體中文 1,041 [1,019–1,063] Percentile 1 · #220 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 960 ±16 1,024 ±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: English 1,087 [1,078–1,096] Percentile 10 · #245 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,107 ±4.87 1,079 ±5 93.5% 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

History

fedi-index by month
The same in a table
Model 2026-10
zephyr-7b-beta 1,082 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: zephyr-7b-beta

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