Skip to content
fedi.software

GPT-5.3 Chat (latest) in fedi-index

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

fedi-index · General

B 1,372 ±2

Percentile 54 · #136 · provisional

People prefer / Solves tasks

1,382 / —

Human votes vs benchmarks, Elo-eq

Value

-45

$4.81 per 1M tokens, median of 6 providers · vendor $4.81

On your hardware

Closed weights: cloud only.

Score decomposition

B General 1,372 [1,369–1,374] Percentile 54 · #136 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,389 ±2.15 1,389 ±2 25.8% 2026-10-02 —
LMArena text/hard_prompts 1,399 ±2.64 1,394 ±2 25.6% 2026-10-02 —
LMArena text/instruction_following 1,379 ±3.30 1,393 ±3 12.6% 2026-10-02 —
LMArena text/expert 1,397 ±5.70 1,392 ±5 12.0% 2026-10-02 —
LMArena text/longer_query 1,396 ±3.16 1,396 ±3 6.3% 2026-10-02 —
LMArena text/multi_turn 1,412 ±4.07 1,410 ±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 1,370 [1,364–1,375] Percentile 53 · #132 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,407 ±3.58 1,388 ±3 90.2% 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 Math 1,359 [1,348–1,370] Percentile 48 · #133 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,389 ±6.64 1,382 ±6 87.1% 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 Writing 1,376 [1,371–1,382] Percentile 56 · #118 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,356 ±4.42 1,384 ±4 61.5% 2026-10-02 —
LMArena text/longer_query 1,396 ±3.16 1,396 ±3 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

B Language: 繁體中文 1,387 [1,378–1,397] Percentile 51 · #109 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,433 ±6.55 1,400 ±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

C Language: Deutsch 1,372 [1,352–1,392] Percentile 41 · #85 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,384 ±13 1,388 ±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

B Language: English 1,367 [1,361–1,372] Percentile 53 · #130 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,390 ±2.95 1,377 ±3 93.9% 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: Español 1,358 [1,344–1,374] Percentile 24 · #113 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,370 ±8.96 1,371 ±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

C Language: Français 1,370 [1,354–1,387] Percentile 29 · #96 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,398 ±9.95 1,384 ±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 [1,360–1,398] Percentile 37 · #86 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,348 ±14 1,395 ±10 91.6% 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: Polski 1,388 [1,371–1,405] Percentile 36 · #84 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,396 ±11 1,404 ±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

B Language: Русский 1,394 [1,386–1,402] Percentile 58 · #99 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,400 ±4.92 1,407 ±4 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
GPT-5.3 Chat (latest) 1,372 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: GPT-5.3 Chat (latest)

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