Skip to content
fedi.software

grok-3-preview-02-24 in fedi-index

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

fedi-index · General

B 1,398 ±3

Percentile 66 · #99 · provisional

People prefer / Solves tasks

1,410 / —

Human votes vs benchmarks, Elo-eq

Value

-24

$6 per 1M tokens, median of 1 providers

On your hardware

Closed weights: cloud only.

Score decomposition

B General 1,398 [1,395–1,400] Percentile 66 · #99 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,426 ±2.20 1,426 ±2 26.1% 2026-10-02 —
LMArena text/hard_prompts 1,433 ±3.24 1,425 ±3 25.7% 2026-10-02 —
LMArena text/instruction_following 1,409 ±3.36 1,421 ±3 12.8% 2026-10-02 —
LMArena text/expert 1,420 ±7.68 1,411 ±6 11.4% 2026-10-02 —
LMArena text/longer_query 1,439 ±4.42 1,437 ±4 6.2% 2026-10-02 —
LMArena text/multi_turn 1,424 ±4.58 1,421 ±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 1,389 [1,382–1,396] Percentile 62 · #107 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,432 ±4.24 1,410 ±4 90.0% 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

C Math 1,361 [1,352–1,371] Percentile 49 · #130 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,390 ±5.66 1,384 ±5 87.6% 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

B Writing 1,426 [1,419–1,432] Percentile 77 · #62 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,414 ±4.66 1,443 ±5 61.9% 2026-10-02 —
LMArena text/longer_query 1,439 ±4.42 1,437 ±4 31.7% 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 1,398 [1,388–1,408] Percentile 55 · #100 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,448 ±6.96 1,412 ±6 93.4% 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: Deutsch 1,411 [1,394–1,428] Percentile 65 · #51 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,431 ±11 1,429 ±10 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: English 1,414 [1,409–1,420] Percentile 69 · #85 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,438 ±2.78 1,428 ±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

C Language: Español 1,392 [1,367–1,417] Percentile 44 · #84 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,417 ±15 1,414 ±14 89.4% 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: Français 1,416 [1,391–1,442] Percentile 58 · #57 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,460 ±16 1,442 ±14 89.3% 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: 日本語 1,402 [1,387–1,416] Percentile 54 · #60 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,388 ±11 1,418 ±8 92.6% 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

B Language: 한국어 1,397 [1,378–1,416] Percentile 51 · #67 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,374 ±14 1,416 ±11 91.5% 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

B Language: Polski 1,419 [1,407–1,431] Percentile 65 · #46 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,433 ±7.86 1,435 ±7 93.0% 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 1,407 [1,398–1,416] Percentile 64 · #84 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,416 ±5.67 1,421 ±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

History

fedi-index by month
The same in a table
Model 2026-10
grok-3-preview-02-24 1,398 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-3-preview-02-24

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