Skip to content
fedi.software

Qwen3.6 Max Preview in fedi-index

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

fedi-index · General

A 1,424 ±7

Percentile 80 · #59

People prefer / Solves tasks

1,432 / 1,414

Human votes vs benchmarks, Elo-eq

Value

+20

$2.92 per 1M tokens, median of 12 providers · vendor $2.92

On your hardware

Closed weights: cloud only.

Score decomposition

A General 1,424 [1,418–1,431] Percentile 80 · #59 · components: 9
Source Component Value Elo-eq Weight Measured Variant
Epoch AI eci 149 ±1.21 1,432 ±10 33.3% 2026-04-20 —
LMArena text/overall 1,447 ±4.17 1,447 ±4 16.5% 2026-10-02 —
LMArena text/hard_prompts 1,457 ±5.19 1,446 ±5 16.2% 2026-10-02 —
LMArena text/instruction_following 1,438 ±7.02 1,449 ±7 7.4% 2026-10-02 —
LMArena text/expert 1,477 ±13 1,457 ±10 6.0% 2026-10-02 —
Epoch AI simpleqa_verified decay ×0.61 0.52 ±0.02 1,461 ±5 4.3% 2026-08-27 —
LMArena text/longer_query 1,458 ±6.63 1,454 ±6 3.8% 2026-10-02 —
LMArena text/multi_turn 1,456 ±10 1,450 ±9 3.2% 2026-10-02 —
Epoch AI gpqa_diamond decay ×0.26 0.87 ±0.02 1,446 ±11 1.3% 2026-08-07 —

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

A Coding 1,427 [1,421–1,432] Percentile 82 · #52 · components: 3
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,482 ±6.38 1,444 ±2 41.7% 2026-10-01 —
LMArena text/coding 1,472 ±7.72 1,445 ±7 34.9% 2026-10-02 —
Epoch AI swe_bench_verified decay ×0.37 0.77 ±0.02 1,466 ±10 14.5% 2026-05-28 —

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 1,407 [1,384–1,429] Percentile 75 · #64 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,465 ±15 1,455 ±14 80.8% 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

A Writing 1,439 [1,426–1,452] Percentile 83 · #46 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,435 ±11 1,464 ±12 54.5% 2026-10-02 —
LMArena text/longer_query 1,458 ±6.63 1,454 ±6 37.3% 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

B Language: 繁體中文 1,419 [1,397–1,442] Percentile 72 · #63 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,486 ±16 1,442 ±13 90.4% 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 1,427 [1,415–1,438] Percentile 79 · #59 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,452 ±6.10 1,443 ±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

B Language: Русский 1,425 [1,405–1,445] Percentile 77 · #54 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,445 ±13 1,447 ±11 91.0% 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

History

fedi-index by month
The same in a table
Model 2026-10
Qwen3.6 Max Preview 1,424 A

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: Qwen3.6 Max Preview

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