Skip to content
fedi.software

starling-lm-7b-beta in fedi-index

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

fedi-index · General

D 1,142 ±4

Percentile 18 · #242 · provisional

People prefer / Solves tasks

1,138 / —

Human votes vs benchmarks, Elo-eq

Value

—

On your hardware

Closed weights: cloud only.

Score decomposition

D General 1,142 [1,138–1,146] Percentile 18 · #242 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,132 ±3.72 1,132 ±4 26.7% 2026-10-02 —
LMArena text/hard_prompts 1,118 ±5.43 1,141 ±5 25.6% 2026-10-02 —
LMArena text/instruction_following 1,096 ±5.09 1,125 ±5 12.8% 2026-10-02 —
LMArena text/expert 1,084 ±10 1,136 ±8 10.9% 2026-10-02 —
LMArena text/multi_turn 1,111 ±7.64 1,137 ±7 5.8% 2026-10-02 —
LMArena text/longer_query 1,104 ±7.83 1,124 ±7 5.7% 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

D Coding 1,157 [1,147–1,167] Percentile 17 · #231 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,143 ±6.47 1,152 ±6 89.3% 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 Math 1,138 [1,127–1,150] Percentile 15 · #218 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,124 ±7.05 1,129 ±7 86.9% 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 1,126 [1,116–1,136] Percentile 15 · #229 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,096 ±7.32 1,118 ±8 61.7% 2026-10-02 —
LMArena text/longer_query 1,104 ±7.83 1,124 ±7 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 1,150 [1,139–1,161] Percentile 11 · #198 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,114 ±7.55 1,147 ±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

D Language: Deutsch 1,135 [1,111–1,159] Percentile 9 · #131 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,081 ±16 1,127 ±14 89.8% 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: English 1,141 [1,132–1,150] Percentile 20 · #220 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,162 ±4.51 1,136 ±5 93.6% 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 1,144 [1,131–1,158] Percentile 9 · #213 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,103 ±8.25 1,140 ±7 92.8% 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

History

fedi-index by month
The same in a table
Model 2026-10
starling-lm-7b-beta 1,142 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: starling-lm-7b-beta

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