Skip to content
fedi.software

GPT-6.1 Sol in fedi-index

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

fedi-index · General

S 1,465 ±9

Percentile 95 · #15

People prefer / Solves tasks

1,437 / 1,513

Human votes vs benchmarks, Elo-eq

Value

+53

$4 per 1M tokens, median of 18 providers · vendor $4

On your hardware

Closed weights: cloud only.

Score decomposition

S General 1,465 [1,456–1,474] Percentile 95 · #15 · components: 8
Source Component Value Elo-eq Weight Measured Variant
Epoch AI eci 166 ±1.97 1,565 ±16 26.2% 2026-09-29 —
LMArena text/overall 1,446 ±5.37 1,446 ±5 21.4% 2026-10-02 max
LMArena text/hard_prompts 1,466 ±7.05 1,455 ±6 20.5% 2026-10-02 max
LMArena text/instruction_following 1,469 ±9.74 1,478 ±9 8.8% 2026-10-02 max
Epoch AI simpleqa_verified decay ×0.61 0.74 ±0.01 1,530 ±4 5.1% 2026-09-29 max
LMArena text/longer_query 1,465 ±8.71 1,461 ±8 4.7% 2026-10-02 max
LMArena text/multi_turn 1,453 ±15 1,447 ±14 3.3% 2026-10-02 max
Epoch AI gpqa_diamond outlier decay ×0.26 0.95 ±0.01 1,483 ±6 1.0% 2026-09-29 max

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 1,464 [1,458–1,469] Percentile 98 · #7 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,758 ±8.59 1,523 ±2 64.4% 2026-10-01 max
LMArena text/coding outlier 1,482 ±12 1,454 ±11 21.8% 2026-10-02 max

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 Agents 1,475 [1,463–1,487] Percentile 91 · #6 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena agent/overall 0.11 ±0.01 1,509 ±7 88.9% 2026-10-02 max

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 1,491 [1,487–1,494] Percentile 100 · #1 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
Epoch AI frontiermath_t1_3 decay ×0.86 0.94 ±0.01 1,524 ±3 57.2% 2026-09-29 max
Epoch AI frontiermath_t4 decay ×0.86 1.00 ±0.00 1,552 ±0 29.8% 2026-09-29 max

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 1,442 [1,431–1,454] Percentile 78 · #26 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena vision/overall 1,285 ±8.12 1,461 ±7 75.6% 2026-10-02 max
LMArena vision/ocr 1,289 ±9.41 1,461 ±9 17.1% 2026-10-02 max

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 1,437 [1,423–1,451] Percentile 82 · #48 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,429 ±12 1,458 ±12 55.8% 2026-10-02 max
LMArena text/longer_query 1,465 ±8.71 1,461 ±8 35.6% 2026-10-02 max

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 1,426 [1,410–1,442] Percentile 78 · #61 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,454 ±8.61 1,445 ±9 92.1% 2026-10-02 max

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 1,424 [1,401–1,446] Percentile 75 · #59 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,445 ±14 1,447 ±13 90.3% 2026-10-02 max

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
GPT-6.1 Sol 1,465 S

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: GPT-6.1 Sol

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