Grok 4.5 in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
A 1,440 ±5
Percentile 87 · #38
People prefer / Solves tasks
1,436 / 1,446
Human votes vs benchmarks, Elo-eq
Value
+35
$3 per 1M tokens, median of 23 providers · vendor $3
On your hardware
Closed weights: cloud only.
Score decomposition
A General 1,440 [1,435–1,445] Percentile 87 · #38 · components: 9
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
154 ±0.87 | 1,469 ±7 | 34.3% | 2026-07-08 | — |
| LMArena |
text/overall
|
1,448 ±2.21 | 1,448 ±2 | 15.5% | 2026-10-02 | — |
| LMArena |
text/hard_prompts
|
1,462 ±2.57 | 1,451 ±2 | 15.4% | 2026-10-02 | — |
| LMArena |
text/instruction_following
|
1,445 ±3.11 | 1,456 ±3 | 7.6% | 2026-10-02 | — |
| LMArena |
text/expert
|
1,466 ±4.73 | 1,449 ±4 | 7.4% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,462 ±2.90 | 1,459 ±3 | 3.8% | 2026-10-02 | — |
| Epoch AI |
simpleqa_verified
decay ×0.61 |
0.48 ±0.02 | 1,449 ±5 | 3.8% | 2026-08-27 | high |
| LMArena |
text/multi_turn
|
1,456 ±4.11 | 1,450 ±4 | 3.7% | 2026-10-02 | — |
| Epoch AI |
gpqa_diamond
decay ×0.26 |
0.93 ±0.01 | 1,474 ±7 | 1.5% | 2026-07-08 | 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 1,432 [1,429–1,434] Percentile 84 · #46 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
webdev/overall
|
1,552 ±3.38 | 1,464 ±1 | 46.0% | 2026-10-01 | — |
| LMArena |
text/coding
|
1,474 ±3.40 | 1,448 ±3 | 44.4% | 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 Agents 1,451 [1,446–1,457] Percentile 75 · #15 · components: 2
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
agent/overall
|
0.01 ±0.01 | 1,461 ±2 | 66.6% | 2026-10-02 | — |
| τ²-bench (Sierra) |
tau2/banking_knowledge
decay ×0.76 |
0.48 ±0.05 | 1,494 ±8 | 26.3% | 2026-07-24 | 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 Math 1,427 [1,419–1,436] Percentile 84 · #41 · components: 3
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,459 ±7.25 | 1,450 ±7 | 50.6% | 2026-10-02 | — |
| Epoch AI |
frontiermath_t1_3
decay ×0.86 |
0.57 ±0.03 | 1,445 ±6 | 30.0% | 2026-07-09 | high |
| Epoch AI |
frontiermath_t4
decay ×0.86 |
0.24 ±0.07 | 1,432 ±11 | 11.6% | 2026-07-09 | 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 Vision 1,449 [1,444–1,453] Percentile 82 · #21 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
vision/overall
|
1,288 ±3.74 | 1,464 ±3 | 63.4% | 2026-10-02 | — |
| LMArena |
vision/ocr
|
1,295 ±4.00 | 1,466 ±4 | 15.6% | 2026-10-02 | — |
| LMArena |
vision/diagram
|
1,307 ±5.96 | 1,470 ±5 | 14.8% | 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 Search 1,439 [1,433–1,446] Percentile 78 · #8 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
search/overall
|
1,213 ±3.50 | 1,455 ±4 | 93.8% | 2026-08-24 | — |
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 1,450 [1,445–1,455] Percentile 86 · #38 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,441 ±3.90 | 1,470 ±4 | 61.8% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,462 ±2.90 | 1,459 ±3 | 32.0% | 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
A Language: 繁體中文 1,434 [1,425–1,442] Percentile 80 · #45 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,495 ±5.77 | 1,449 ±5 | 93.6% | 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: Deutsch 1,423 [1,405–1,442] Percentile 73 · #39 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,448 ±12 | 1,443 ±10 | 91.6% | 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
A Language: English 1,430 [1,424–1,436] Percentile 82 · #49 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,454 ±2.94 | 1,445 ±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
B Language: Español 1,426 [1,410–1,442] Percentile 72 · #42 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,450 ±9.54 | 1,445 ±9 | 92.2% | 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 Language: Français 1,420 [1,404–1,435] Percentile 62 · #52 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,456 ±9.36 | 1,438 ±9 | 92.3% | 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 1,426 [1,407–1,445] Percentile 78 · #29 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,427 ±14 | 1,447 ±11 | 91.4% | 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: 한국어 1,419 [1,403–1,435] Percentile 68 · #44 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,402 ±12 | 1,437 ±9 | 92.2% | 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 Language: Polski 1,418 [1,399–1,436] Percentile 62 · #51 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,435 ±12 | 1,437 ±10 | 91.6% | 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
A Language: Русский 1,434 [1,426–1,442] Percentile 81 · #46 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,447 ±4.71 | 1,449 ±4 | 93.7% | 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
History
The same in a table
| Model | 2026-10 |
|---|---|
| Grok 4.5 | 1,440 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.