Skip to content
fedi.software

Qwen3 Coder 480B A35B in fedi-index

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

fedi-index · General

C 1,344 ±3

Percentile 46 · #160 · provisional

People prefer / Solves tasks

1,353 / —

Human votes vs benchmarks, Elo-eq

Value

-22

$0.615 per 1M tokens, median of 21 providers · vendor $3

On your hardware

Does not fit the reference setups.

Score decomposition

C General 1,344 [1,342–1,347] Percentile 46 · #160 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,356 ±2.55 1,356 ±3 26.1% 2026-10-02 —
LMArena text/hard_prompts 1,372 ±3.28 1,370 ±3 25.9% 2026-10-02 —
LMArena text/instruction_following 1,356 ±4.03 1,371 ±4 12.6% 2026-10-02 —
LMArena text/expert 1,337 ±8.33 1,343 ±7 11.2% 2026-10-02 —
LMArena text/longer_query 1,378 ±4.44 1,380 ±4 6.2% 2026-10-02 —
LMArena text/multi_turn 1,365 ±4.91 1,368 ±4 6.2% 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

B Coding 1,371 [1,367–1,374] Percentile 53 · #131 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,274 ±3.95 1,385 ±1 46.6% 2026-10-01 —
LMArena text/coding 1,412 ±4.60 1,392 ±4 43.6% 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

C Math 1,337 [1,325–1,350] Percentile 42 · #148 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,364 ±7.58 1,358 ±7 86.6% 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

C Writing 1,357 [1,349–1,364] Percentile 49 · #136 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,333 ±5.47 1,360 ±6 61.1% 2026-10-02 —
LMArena text/longer_query 1,378 ±4.44 1,380 ±4 32.3% 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

C Language: 繁體中文 1,330 [1,319–1,341] Percentile 34 · #148 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,356 ±7.48 1,339 ±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

C Language: Deutsch 1,325 [1,305–1,345] Percentile 28 · #104 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,325 ±13 1,337 ±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

C Language: English 1,336 [1,330–1,343] Percentile 42 · #159 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,359 ±3.33 1,345 ±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: Español 1,348 [1,328–1,367] Percentile 22 · #117 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,360 ±12 1,362 ±11 91.2% 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

C Language: Français 1,362 [1,338–1,387] Percentile 27 · #99 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,394 ±15 1,381 ±14 89.5% 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: 日本語 1,347 [1,333–1,362] Percentile 30 · #90 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,309 ±11 1,359 ±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

C Language: 한국어 1,351 [1,333–1,368] Percentile 28 · #98 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,306 ±13 1,364 ±10 91.8% 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: Polski 1,349 [1,337–1,361] Percentile 18 · #108 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,344 ±7.71 1,360 ±6 93.1% 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

C Language: Русский 1,365 [1,353–1,377] Percentile 45 · #129 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,367 ±7.25 1,377 ±6 93.1% 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
Qwen3 Coder 480B A35B 1,344 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: Qwen3 Coder 480B A35B

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