Skip to content
fedi.software

Grok 4 Fast (Search) in fedi-index

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

fedi-index · General

B 1,381 ±4

Percentile 56 · #131 · provisional

People prefer / Solves tasks

1,393 / —

Human votes vs benchmarks, Elo-eq

Value

—

On your hardware

Closed weights: cloud only.

Score decomposition

B General 1,381 [1,376–1,385] Percentile 56 · #131 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,407 ±3.96 1,407 ±4 27.0% 2026-11-11 non-reasoning
LMArena text/hard_prompts 1,412 ±5.45 1,406 ±5 26.0% 2026-11-11 non-reasoning
LMArena text/instruction_following 1,387 ±7.42 1,400 ±7 11.8% 2026-11-11 non-reasoning
LMArena text/expert 1,413 ±9.69 1,405 ±8 11.3% 2026-11-11 reasoning
LMArena text/longer_query 1,414 ±7.99 1,413 ±8 5.8% 2026-11-11 non-reasoning
LMArena text/multi_turn 1,414 ±9.06 1,411 ±8 5.5% 2026-11-11 non-reasoning

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 1,343 [1,337–1,349] Percentile 39 · #176 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,159 ±14 1,352 ±4 59.1% 2026-11-15 reasoning
LMArena text/coding outlier 1,429 ±8.54 1,408 ±8 25.1% 2026-11-11 non-reasoning

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 1,370 [1,348–1,393] Percentile 50 · #129 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,419 ±15 1,411 ±14 80.6% 2026-11-11 non-reasoning

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 Search 1,399 [1,394–1,404] Percentile 23 · #28 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena search/overall 1,171 ±2.41 1,412 ±2 94.0% 2026-11-09 search

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 1,397 [1,384–1,410] Percentile 62 · #100 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,386 ±10 1,415 ±10 57.7% 2026-11-11 non-reasoning
LMArena text/longer_query 1,414 ±7.99 1,413 ±8 34.3% 2026-11-11 non-reasoning

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 1,400 [1,379–1,420] Percentile 55 · #99 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,457 ±14 1,419 ±12 91.0% 2026-11-11 non-reasoning

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 1,368 [1,344–1,392] Percentile 39 · #89 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,382 ±15 1,387 ±13 89.9% 2026-11-11 reasoning

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 1,389 [1,378–1,400] Percentile 57 · #117 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,414 ±5.80 1,403 ±6 93.2% 2026-11-11 non-reasoning

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 1,385 [1,363–1,406] Percentile 40 · #83 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,358 ±16 1,404 ±12 90.7% 2026-11-11 reasoning

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 1,407 [1,384–1,431] Percentile 45 · #72 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,427 ±16 1,430 ±13 89.9% 2026-11-11 non-reasoning

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 1,383 [1,370–1,396] Percentile 49 · #117 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,389 ±7.68 1,397 ±7 93.0% 2026-11-11 reasoning

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
Grok 4 Fast (Search) 1,381 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: Grok 4 Fast (Search)

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