Skip to content
fedi.software

Grok 4 (Search) in fedi-index

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

fedi-index · General

B 1,383 ±2

Percentile 58 · #125 · provisional

People prefer / Solves tasks

1,394 / —

Human votes vs benchmarks, Elo-eq

Value

-46

$6 per 1M tokens, median of 6 providers

On your hardware

Closed weights: cloud only.

Score decomposition

B General 1,383 [1,381–1,385] Percentile 58 · #125 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,412 ±2.01 1,412 ±2 25.9% 2026-11-11 —
LMArena text/hard_prompts 1,409 ±2.56 1,403 ±2 25.7% 2026-11-11 —
LMArena text/instruction_following 1,387 ±3.15 1,401 ±3 12.7% 2026-11-11 —
LMArena text/expert 1,415 ±6.67 1,407 ±6 11.6% 2026-11-11 —
LMArena text/longer_query 1,408 ±3.39 1,408 ±3 6.3% 2026-11-11 —
LMArena text/multi_turn 1,416 ±3.97 1,414 ±4 6.2% 2026-11-11 —

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 1,367 [1,362–1,373] Percentile 50 · #145 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,409 ±3.53 1,389 ±3 88.5% 2026-11-11 —

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 Math 1,388 [1,377–1,398] Percentile 58 · #108 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,422 ±6.32 1,414 ±6 87.3% 2026-11-11 —

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 Vision 1,376 [1,371–1,380] Percentile 34 · #73 · components: 3 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena vision/overall 1,210 ±3.72 1,394 ±3 63.9% 2026-11-09 —
LMArena vision/ocr 1,195 ±4.67 1,370 ±4 15.4% 2026-11-09 —
LMArena vision/diagram 1,203 ±6.91 1,377 ±6 14.4% 2026-11-09 —

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 1,371 [1,365–1,377] Percentile 6 · #34 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena search/overall 1,142 ±3.15 1,382 ±3 93.9% 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 1,405 [1,400–1,411] Percentile 67 · #88 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,396 ±4.30 1,424 ±4 61.7% 2026-11-11 —
LMArena text/longer_query 1,408 ±3.39 1,408 ±3 32.0% 2026-11-11 —

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 1,383 [1,375–1,392] Percentile 46 · #119 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,427 ±5.81 1,396 ±5 93.6% 2026-11-11 —

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: Deutsch 1,410 [1,394–1,426] Percentile 60 · #59 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,429 ±10 1,427 ±9 92.2% 2026-11-11 —

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 1,395 [1,390–1,400] Percentile 59 · #110 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,418 ±2.65 1,407 ±3 93.9% 2026-11-11 —

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: Español 1,400 [1,384–1,417] Percentile 40 · #91 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,420 ±9.74 1,417 ±9 92.2% 2026-11-11 —

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 1,385 [1,364–1,405] Percentile 35 · #93 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,417 ±13 1,403 ±12 90.9% 2026-11-11 —

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: 日本語 1,406 [1,392–1,420] Percentile 55 · #58 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,393 ±10 1,422 ±8 92.6% 2026-11-11 —

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: 한국어 1,401 [1,386–1,416] Percentile 50 · #69 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,376 ±11 1,417 ±8 92.5% 2026-11-11 —

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: Polski 1,417 [1,408–1,426] Percentile 58 · #56 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,429 ±6.03 1,432 ±5 93.5% 2026-11-11 —

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 1,402 [1,392–1,411] Percentile 60 · #93 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,409 ±5.57 1,415 ±5 93.5% 2026-11-11 —

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 (Search) 1,383 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 (Search)

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