Skip to content
fedi.software

qwen1.5-4b-chat in fedi-index

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

fedi-index · General

D 1,038 ±5

Percentile 3 · #283 · provisional

People prefer / Solves tasks

1,026 / —

Human votes vs benchmarks, Elo-eq

Value

—

On your hardware

Closed weights: cloud only.

Score decomposition

D General 1,038 [1,032–1,043] Percentile 3 · #283 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 997 ±4.70 997 ±5 28.0% 2026-10-02 —
LMArena text/hard_prompts 976 ±7.07 1,012 ±6 26.0% 2026-10-02 —
LMArena text/instruction_following 979 ±6.54 1,014 ±6 13.1% 2026-10-02 —
LMArena text/expert 980 ±16 1,051 ±13 8.8% 2026-10-02 —
LMArena text/multi_turn 977 ±9.56 1,015 ±9 5.8% 2026-10-02 —
LMArena text/longer_query 988 ±12 1,015 ±11 4.9% 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 Coding 1,045 [1,031–1,058] Percentile 4 · #268 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 999 ±8.76 1,024 ±8 88.3% 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 1,058 [1,044–1,073] Percentile 5 · #243 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,026 ±8.99 1,034 ±9 85.7% 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 Writing 1,014 [1,000–1,028] Percentile 3 · #259 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 969 ±9.56 988 ±10 62.9% 2026-10-02 —
LMArena text/longer_query 988 ±12 1,015 ±11 28.7% 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 Language: 繁體中文 1,085 [1,070–1,099] Percentile 4 · #213 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,024 ±10 1,075 ±8 92.6% 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: English 996 [986–1,007] Percentile 3 · #264 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,015 ±5.52 981 ±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: Русский 1,022 [1,002–1,042] Percentile 0 · #233 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 952 ±13 1,004 ±11 91.1% 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

History

fedi-index by month
The same in a table
Model 2026-10
qwen1.5-4b-chat 1,038 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: qwen1.5-4b-chat

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