Grok 4.6 in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
A 1,437 ±6
Percentile 85 · #44
People prefer / Solves tasks
1,422 / 1,460
Human votes vs benchmarks, Elo-eq
Value
+32
$3 per 1M tokens, median of 26 providers · vendor $3
On your hardware
Closed weights: cloud only.
Score decomposition
A General 1,437 [1,431–1,443] Percentile 85 · #44 · components: 9
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
156 ±1.14 | 1,489 ±9 | 32.0% | 2026-08-12 | — |
| LMArena |
text/overall
|
1,427 ±2.62 | 1,427 ±3 | 16.1% | 2026-10-02 | high |
| LMArena |
text/hard_prompts
|
1,446 ±3.03 | 1,436 ±3 | 16.0% | 2026-10-02 | high |
| LMArena |
text/instruction_following
|
1,430 ±3.72 | 1,442 ±4 | 7.8% | 2026-10-02 | high |
| LMArena |
text/expert
|
1,465 ±5.97 | 1,447 ±5 | 7.5% | 2026-10-02 | high |
| Epoch AI |
simpleqa_verified
decay ×0.61 |
0.49 ±0.02 | 1,452 ±5 | 4.0% | 2026-08-27 | high |
| LMArena |
text/longer_query
|
1,453 ±3.44 | 1,450 ±3 | 4.0% | 2026-10-02 | high |
| LMArena |
text/multi_turn
|
1,423 ±5.26 | 1,420 ±5 | 3.8% | 2026-10-02 | high |
| Epoch AI |
gpqa_diamond
decay ×0.26 |
0.94 ±0.01 | 1,477 ±7 | 1.6% | 2026-08-12 | 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 Coding 1,436 [1,433–1,440] Percentile 87 · #37 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
webdev/overall
|
1,620 ±4.00 | 1,484 ±1 | 46.4% | 2026-10-01 | high |
| LMArena |
text/coding
|
1,463 ±4.19 | 1,438 ±4 | 43.9% | 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
B Agents 1,437 [1,432–1,441] Percentile 56 · #26 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
agent/overall
|
0.01 ±0.01 | 1,462 ±2 | 90.3% | 2026-10-02 | xhigh |
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 Math 1,416 [1,406–1,425] Percentile 81 · #50 · components: 3
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,421 ±8.97 | 1,413 ±9 | 48.6% | 2026-10-02 | high |
| Epoch AI |
frontiermath_t1_3
decay ×0.86 |
0.66 ±0.03 | 1,464 ±6 | 31.9% | 2026-08-14 | xhigh |
| Epoch AI |
frontiermath_t4
decay ×0.86 |
0.32 ±0.07 | 1,443 ±12 | 11.5% | 2026-08-14 | xhigh |
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 Vision 1,429 [1,423–1,435] Percentile 66 · #39 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
vision/overall
|
1,265 ±4.63 | 1,443 ±4 | 64.0% | 2026-10-02 | high |
| LMArena |
vision/ocr
|
1,273 ±5.20 | 1,446 ±5 | 15.5% | 2026-10-02 | high |
| LMArena |
vision/diagram
|
1,287 ±8.17 | 1,452 ±7 | 14.0% | 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
A Writing 1,440 [1,434–1,446] Percentile 84 · #45 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,430 ±4.69 | 1,459 ±5 | 61.3% | 2026-10-02 | high |
| LMArena |
text/longer_query
|
1,453 ±3.44 | 1,450 ±3 | 32.3% | 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
B Language: 繁體中文 1,421 [1,410–1,432] Percentile 74 · #59 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,479 ±7.21 | 1,437 ±6 | 93.3% | 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
B Language: Deutsch 1,402 [1,379–1,425] Percentile 58 · #61 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,425 ±15 | 1,423 ±13 | 90.1% | 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
B Language: English 1,407 [1,400–1,414] Percentile 66 · #93 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,431 ±3.50 | 1,420 ±4 | 93.8% | 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: Español 1,387 [1,367–1,407] Percentile 38 · #92 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,407 ±12 | 1,405 ±11 | 91.1% | 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
B Language: Français 1,420 [1,400–1,440] Percentile 64 · #49 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,460 ±12 | 1,442 ±11 | 91.0% | 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: 日本語 1,389 [1,368–1,411] Percentile 45 · #71 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,376 ±16 | 1,409 ±12 | 90.5% | 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
B Language: 한국어 1,414 [1,394–1,433] Percentile 64 · #49 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,398 ±14 | 1,434 ±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
B Language: Polski 1,408 [1,384–1,431] Percentile 52 · #63 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,428 ±16 | 1,431 ±13 | 89.9% | 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
B Language: Русский 1,415 [1,405–1,424] Percentile 69 · #73 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,425 ±5.75 | 1,430 ±5 | 93.5% | 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
The same in a table
| Model | 2026-10 |
|---|---|
| Grok 4.6 | 1,437 A |
Badge
Show the model's fedi-index on your site or in a README. The badge updates itself; it links to this page.
Data CC BY 4.0: keep the link to fedi.software.