Skip to content
fedi.software

GPT-6.1 Terra in fedi-index

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

fedi-index · General

A 1,449 ±13

Percentile 87 · #40 · provisional

People prefer / Solves tasks

— / 1,490

Human votes vs benchmarks, Elo-eq

Value

+54

$2 per 1M tokens, median of 28 providers · vendor $2

On your hardware

Closed weights: cloud only.

Score decomposition

A General 1,449 [1,437–1,462] Percentile 87 · #40 · components: 3 provisional
Source Component Value Elo-eq Weight Measured Variant
Epoch AI eci outlier 168 ±1.59 1,577 ±13 50.5% 2026-10-26 —
Epoch AI simpleqa_verified decay ×0.58 0.57 ±0.02 1,477 ±5 14.9% 2026-10-26 max
Epoch AI gpqa_diamond decay ×0.25 0.94 ±0.01 1,478 ±7 5.9% 2026-10-26 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

S Coding 1,476 [1,473–1,478] Percentile 97 · #10 · components: 2
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,681 ±5.26 1,501 ±2 67.8% 2026-11-15 max
Terminal-Bench tbench/terminal-bench decay ×0.90 0.53 ±0.02 1,501 ±3 23.7% 2026-10-26 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 1,481 [1,476–1,485] Percentile 90 · #9 · components: 2
Source Component Value Elo-eq Weight Measured Variant
LMArena agent/overall 0.11 ±0.01 1,510 ±3 67.1% 2026-11-15 max
Terminal-Bench tbench/terminal-bench decay ×0.90 0.53 ±0.02 1,501 ±3 24.2% 2026-10-26 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 1,478 [1,473–1,484] Percentile 97 · #9 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
Epoch AI frontiermath_t1_3 decay ×0.82 0.91 ±0.02 1,518 ±4 59.2% 2026-10-26 max
Epoch AI frontiermath_t4 decay ×0.82 0.93 ±0.04 1,540 ±7 26.4% 2026-10-26 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 2026-11
GPT-6.1 Terra 1,449 A

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 Terra

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