Skip to content
fedi.software

GPT-5.6 Sol in fedi-index

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

fedi-index · General

S 1,469 ±7

Percentile 97 · #11

People prefer / Solves tasks

1,454 / 1,493

Human votes vs benchmarks, Elo-eq

Value

+40

$8 per 1M tokens, median of 34 providers · vendor $8

On your hardware

Closed weights: cloud only.

Score decomposition

S General 1,469 [1,462–1,476] Percentile 97 · #11 · components: 9
Source Component Value Elo-eq Weight Measured Variant
Epoch AI eci 162 ±1.49 1,530 ±12 28.2% 2026-07-09 —
LMArena text/overall 1,456 ±2.26 1,456 ±2 16.9% 2026-10-02 xhigh
LMArena text/hard_prompts 1,482 ±2.62 1,469 ±2 16.9% 2026-10-02 xhigh
LMArena text/instruction_following 1,482 ±3.23 1,490 ±3 8.3% 2026-10-02 xhigh
LMArena text/expert 1,518 ±5.00 1,491 ±4 8.1% 2026-10-02 xhigh
Epoch AI simpleqa_verified decay ×0.61 0.70 ±0.01 1,516 ±5 4.2% 2026-08-10 max
LMArena text/longer_query 1,480 ±3.01 1,475 ±3 4.2% 2026-10-02 xhigh
LMArena text/multi_turn 1,459 ±4.27 1,452 ±4 4.1% 2026-10-02 xhigh
Epoch AI gpqa_diamond decay ×0.26 0.93 ±0.02 1,475 ±7 1.6% 2026-07-09 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 1,449 [1,446–1,452] Percentile 92 · #22 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,620 ±3.06 1,484 ±1 46.1% 2026-10-01 xhigh-codex-harness
LMArena text/coding 1,496 ±3.51 1,467 ±3 44.3% 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 Agents 1,467 [1,461–1,473] Percentile 86 · #9 · components: 2
Source Component Value Elo-eq Weight Measured Variant
LMArena agent/overall 0.06 ±0.01 1,486 ±3 66.3% 2026-10-02 xhigh
τ²-bench (Sierra) tau2/banking_knowledge decay ×0.76 0.47 ±0.05 1,492 ±8 26.5% 2026-07-22 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

S Math 1,467 [1,459–1,474] Percentile 96 · #10 · components: 3
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,472 ±7.30 1,462 ±7 48.6% 2026-10-02 xhigh
Epoch AI frontiermath_t1_3 decay ×0.86 0.89 ±0.02 1,514 ±4 31.8% 2026-07-09 max
Epoch AI frontiermath_t4 decay ×0.86 0.83 ±0.06 1,525 ±10 12.2% 2026-07-09 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 [1,438–1,447] Percentile 76 · #28 · components: 3 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena vision/overall 1,279 ±3.84 1,456 ±4 63.4% 2026-10-02 xhigh
LMArena vision/ocr 1,289 ±4.11 1,460 ±4 15.6% 2026-10-02 xhigh
LMArena vision/diagram 1,303 ±6.29 1,466 ±6 14.7% 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 Search 1,482 [1,475–1,490] Percentile 100 · #1 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena search/overall 1,257 ±3.81 1,501 ±4 93.7% 2026-08-24 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 Writing 1,460 [1,454–1,465] Percentile 92 · #23 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,448 ±4.05 1,478 ±4 61.7% 2026-10-02 xhigh
LMArena text/longer_query 1,480 ±3.01 1,475 ±3 32.1% 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 Language: 繁體中文 1,454 [1,445–1,463] Percentile 91 · #20 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,523 ±6.03 1,471 ±5 93.6% 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 Language: Deutsch 1,441 [1,422–1,460] Percentile 86 · #21 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,471 ±13 1,464 ±11 91.3% 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 Language: English 1,434 [1,428–1,439] Percentile 85 · #43 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,457 ±3.01 1,449 ±3 93.9% 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

B Language: Español 1,423 [1,407–1,439] Percentile 70 · #45 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,446 ±9.54 1,441 ±9 92.2% 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

B Language: Français 1,436 [1,420–1,452] Percentile 79 · #29 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,475 ±9.74 1,456 ±9 92.2% 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 Language: 日本語 1,456 [1,437–1,475] Percentile 89 · #15 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,471 ±14 1,480 ±11 91.4% 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 Language: 한국어 1,442 [1,426–1,458] Percentile 86 · #20 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,436 ±12 1,462 ±9 92.2% 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 Language: Polski 1,437 [1,418–1,456] Percentile 82 · #24 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,461 ±13 1,459 ±11 91.3% 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 Language: Русский 1,451 [1,443–1,459] Percentile 89 · #27 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,468 ±4.90 1,468 ±4 93.6% 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

History

fedi-index by month
The same in a table
Model 2026-10
GPT-5.6 Sol 1,469 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-5.6 Sol

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