Skip to content
fedi.software

Gemini 4 Hydrogen in fedi-index

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

fedi-index · General

A 1,467 ±5

Percentile 93 · #21 · provisional

People prefer / Solves tasks

1,486 / —

Human votes vs benchmarks, Elo-eq

Value

+60

$3 per 1M tokens, median of 15 providers · vendor $3

On your hardware

Closed weights: cloud only.

Score decomposition

A General 1,467 [1,463–1,472] Percentile 93 · #21 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,502 ±4.58 1,502 ±5 27.5% 2026-11-11 high
LMArena text/hard_prompts 1,520 ±5.73 1,503 ±5 26.9% 2026-11-11 high
LMArena text/instruction_following 1,519 ±7.67 1,526 ±7 12.2% 2026-11-11 high
LMArena text/expert 1,558 ±15 1,524 ±12 8.9% 2026-11-11 high
LMArena text/longer_query 1,524 ±6.99 1,516 ±6 6.3% 2026-11-11 high
LMArena text/multi_turn 1,485 ±11 1,476 ±10 5.1% 2026-11-11 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

S Coding 1,472 [1,467–1,478] Percentile 96 · #12 · components: 3
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,704 ±7.26 1,508 ±2 36.5% 2026-11-15 high
LMArena text/coding 1,516 ±9.37 1,485 ±8 28.4% 2026-11-11 high
Terminal-Bench tbench/terminal-bench decay ×0.90 0.56 ±0.02 1,507 ±3 25.8% 2026-11-12 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

A Agents 1,492 [1,486–1,499] Percentile 95 · #5 · components: 2
Source Component Value Elo-eq Weight Measured Variant
LMArena agent/overall 0.15 ±0.01 1,527 ±5 65.9% 2026-11-15 high
Terminal-Bench tbench/terminal-bench decay ×0.90 0.56 ±0.02 1,507 ±3 25.0% 2026-11-12 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

A Writing 1,505 [1,493–1,518] Percentile 98 · #6 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,509 ±10 1,540 ±10 56.9% 2026-11-11 high
LMArena text/longer_query 1,524 ±6.99 1,516 ±6 35.2% 2026-11-11 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

A Language: 繁體中文 1,521 [1,497–1,545] Percentile 100 · #1 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,631 ±17 1,557 ±14 89.8% 2026-11-11 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

A Language: English 1,471 [1,456–1,486] Percentile 96 · #12 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,499 ±7.61 1,493 ±8 92.6% 2026-11-11 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

A Language: Русский 1,466 [1,448–1,484] Percentile 92 · #19 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,492 ±11 1,490 ±10 91.7% 2026-11-11 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 2026-11
Gemini 4 Hydrogen 1,467 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: Gemini 4 Hydrogen

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