Skip to content
fedi.software

yi-lightning in fedi-index

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

fedi-index · General

C 1,292 ±3

Percentile 33 · #196 · provisional

People prefer / Solves tasks

1,297 / —

Human votes vs benchmarks, Elo-eq

Value

—

On your hardware

Closed weights: cloud only.

Score decomposition

C General 1,292 [1,289–1,295] Percentile 33 · #196 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,301 ±2.50 1,302 ±2 26.2% 2026-10-02 —
LMArena text/hard_prompts 1,302 ±4.04 1,306 ±4 25.4% 2026-10-02 —
LMArena text/instruction_following 1,278 ±3.52 1,298 ±3 12.8% 2026-10-02 —
LMArena text/expert 1,285 ±7.45 1,301 ±6 11.6% 2026-10-02 —
LMArena text/multi_turn 1,311 ±5.04 1,318 ±5 6.2% 2026-10-02 —
LMArena text/longer_query 1,297 ±5.36 1,304 ±5 6.1% 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 Coding 1,293 [1,285–1,301] Percentile 31 · #191 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,312 ±4.91 1,303 ±4 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

C Math 1,286 [1,277–1,294] Percentile 33 · #172 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,300 ±5.01 1,297 ±5 87.9% 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 Writing 1,299 [1,291–1,306] Percentile 37 · #169 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,280 ±5.30 1,306 ±5 61.9% 2026-10-02 —
LMArena text/longer_query 1,297 ±5.36 1,304 ±5 31.5% 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: 繁體中文 1,303 [1,294–1,312] Percentile 29 · #159 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,320 ±6.30 1,310 ±5 93.5% 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

D Language: Deutsch 1,280 [1,263–1,298] Percentile 23 · #111 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,267 ±11 1,287 ±10 91.9% 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: English 1,307 [1,300–1,313] Percentile 37 · #173 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,330 ±3.33 1,314 ±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 1,307 [1,281–1,333] Percentile 12 · #132 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,315 ±16 1,320 ±15 88.9% 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 Language: 日本語 1,290 [1,274–1,307] Percentile 20 · #103 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,228 ±12 1,298 ±9 92.0% 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

D Language: 한국어 1,270 [1,249–1,292] Percentile 14 · #117 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,192 ±16 1,277 ±12 90.7% 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: Русский 1,272 [1,264–1,280] Percentile 24 · #177 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,255 ±5.08 1,277 ±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

History

fedi-index by month
The same in a table
Model 2026-10
yi-lightning 1,292 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: yi-lightning

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