Skip to content
fedi.software

Qwen3.5-Flash in fedi-index

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

fedi-index · General

B 1,374 ±2

Percentile 54 · #134 · provisional

People prefer / Solves tasks

1,384 / —

Human votes vs benchmarks, Elo-eq

Value

+38

$0.175 per 1M tokens, median of 11 providers · vendor $0.175

On your hardware

Closed weights: cloud only.

Score decomposition

B General 1,374 [1,372–1,376] Percentile 54 · #134 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,397 ±1.86 1,397 ±2 25.6% 2026-10-02 —
LMArena text/hard_prompts 1,403 ±2.24 1,398 ±2 25.5% 2026-10-02 —
LMArena text/instruction_following 1,374 ±2.72 1,388 ±3 12.6% 2026-10-02 —
LMArena text/expert 1,407 ±4.26 1,400 ±4 12.3% 2026-10-02 —
LMArena text/longer_query 1,392 ±2.65 1,393 ±2 6.3% 2026-10-02 —
LMArena text/multi_turn 1,393 ±3.36 1,392 ±3 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 1,367 [1,363–1,370] Percentile 51 · #137 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,412 ±2.88 1,392 ±3 45.3% 2026-10-02 —
LMArena webdev/overall 1,243 ±10 1,376 ±3 45.0% 2026-10-01 —

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 Math 1,376 [1,367–1,384] Percentile 54 · #117 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,407 ±5.41 1,400 ±5 87.7% 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

B Writing 1,367 [1,362–1,372] Percentile 53 · #127 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,343 ±3.54 1,370 ±4 62.0% 2026-10-02 —
LMArena text/longer_query 1,392 ±2.65 1,393 ±2 31.9% 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

B Language: 繁體中文 1,397 [1,389–1,404] Percentile 54 · #103 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,445 ±5.14 1,410 ±4 93.7% 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 Language: Deutsch 1,379 [1,364–1,394] Percentile 44 · #81 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,390 ±9.61 1,393 ±8 92.5% 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

B Language: English 1,381 [1,376–1,386] Percentile 57 · #118 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,405 ±2.51 1,393 ±3 94.0% 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 Language: Español 1,385 [1,372–1,397] Percentile 37 · #94 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,400 ±7.24 1,398 ±7 93.0% 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 Language: Français 1,384 [1,372–1,396] Percentile 37 · #86 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,412 ±7.25 1,398 ±7 93.0% 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 Language: 日本語 1,387 [1,371–1,404] Percentile 43 · #73 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,368 ±12 1,403 ±9 92.1% 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: 한국어 1,379 [1,366–1,393] Percentile 38 · #85 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,345 ±9.83 1,393 ±8 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

C Language: Polski 1,375 [1,362–1,388] Percentile 30 · #92 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,377 ±8.36 1,388 ±7 92.9% 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

C Language: Русский 1,376 [1,370–1,382] Percentile 48 · #122 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,379 ±3.86 1,388 ±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

History

fedi-index by month
The same in a table
Model 2026-10
Qwen3.5-Flash 1,374 B

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: Qwen3.5-Flash

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