Skip to content
fedi.software

GPT-5 Nano in fedi-index

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

fedi-index · General

C 1,323 ±8

Percentile 39 · #178

People prefer / Solves tasks

1,316 / 1,334

Human votes vs benchmarks, Elo-eq

Value

-7

$0.1375 per 1M tokens, median of 25 providers · vendor $0.1375

On your hardware

Closed weights: cloud only.

Score decomposition

C General 1,323 [1,314–1,331] Percentile 39 · #178 · components: 9
Source Component Value Elo-eq Weight Measured Variant
Epoch AI eci 139 ±1.87 1,354 ±15 26.3% 2025-08-07 —
LMArena text/overall 1,320 ±3.50 1,320 ±4 18.7% 2026-10-02 high
LMArena text/hard_prompts 1,329 ±4.95 1,330 ±4 18.1% 2026-10-02 high
LMArena text/instruction_following 1,306 ±6.68 1,324 ±6 8.4% 2026-10-02 high
LMArena text/expert 1,321 ±16 1,330 ±13 5.5% 2026-10-02 high
Epoch AI simpleqa_verified decay ×0.61 0.12 ±0.01 1,334 ±3 5.0% 2026-08-10 high
LMArena text/longer_query 1,312 ±7.36 1,318 ±7 4.1% 2026-10-02 high
LMArena text/multi_turn 1,311 ±7.99 1,318 ±7 4.0% 2026-10-02 high
Epoch AI gpqa_diamond decay ×0.26 0.69 ±0.03 1,363 ±13 1.3% 2025-10-30 high

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 1,323 [1,311–1,334] Percentile 36 · #179 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,351 ±7.45 1,338 ±7 88.9% 2026-10-02 high

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 [1,327–1,348] Percentile 42 · #149 · components: 3
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,318 ±14 1,314 ±13 38.5% 2026-10-02 high
Epoch AI frontiermath_t1_3 decay ×0.86 0.20 ±0.02 1,364 ±5 34.8% 2026-06-12 high
Epoch AI frontiermath_t4 decay ×0.86 0.02 ±0.02 1,396 ±4 18.2% 2026-06-12 high

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 Vision 1,332 [1,324–1,339] Percentile 23 · #87 · components: 3 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena vision/overall 1,159 ±5.64 1,347 ±5 66.7% 2026-10-02 high
LMArena vision/ocr 1,152 ±7.68 1,329 ±7 15.0% 2026-10-02 high
LMArena vision/diagram 1,141 ±13 1,322 ±12 11.4% 2026-10-02 high

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 1,284 [1,272–1,297] Percentile 33 · #179 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,250 ±10 1,275 ±10 57.3% 2026-10-02 high
LMArena text/longer_query 1,312 ±7.36 1,318 ±7 34.8% 2026-10-02 high

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 1,326 [1,307–1,345] Percentile 32 · #153 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,354 ±13 1,338 ±11 91.4% 2026-10-02 high

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 1,298 [1,289–1,308] Percentile 35 · #178 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,321 ±5.04 1,305 ±5 93.4% 2026-10-02 high

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 1,322 [1,303–1,341] Percentile 10 · #118 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,312 ±13 1,333 ±11 91.4% 2026-10-02 high

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 1,304 [1,284–1,323] Percentile 30 · #163 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,296 ±12 1,313 ±11 91.3% 2026-10-02 high

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 Nano 1,323 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: GPT-5 Nano

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