GPT-5.6 Terra in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
A 1,453 ±7
Percentile 91 · #27
People prefer / Solves tasks
1,440 / 1,476
Human votes vs benchmarks, Elo-eq
Value
+38
$4.50 per 1M tokens, median of 33 providers · vendor $4.50
On your hardware
Closed weights: cloud only.
Score decomposition
A General 1,453 [1,446–1,460] Percentile 91 · #27 · components: 9
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
160 ±1.40 | 1,514 ±11 | 29.7% | 2026-07-09 | — |
| LMArena |
text/overall
|
1,446 ±2.23 | 1,446 ±2 | 17.1% | 2026-10-02 | xhigh |
| LMArena |
text/hard_prompts
|
1,468 ±2.60 | 1,456 ±2 | 17.0% | 2026-10-02 | xhigh |
| LMArena |
text/instruction_following
|
1,453 ±3.17 | 1,463 ±3 | 8.4% | 2026-10-02 | xhigh |
| LMArena |
text/expert
|
1,492 ±4.84 | 1,470 ±4 | 8.2% | 2026-10-02 | xhigh |
| LMArena |
text/longer_query
|
1,451 ±2.98 | 1,448 ±3 | 4.2% | 2026-10-02 | xhigh |
| LMArena |
text/multi_turn
|
1,450 ±4.22 | 1,444 ±4 | 4.1% | 2026-10-02 | xhigh |
| Epoch AI |
simpleqa_verified
outlier decay ×0.61 |
0.43 ±0.02 | 1,433 ±5 | 2.1% | 2026-08-10 | max |
| Epoch AI |
gpqa_diamond
decay ×0.26 |
0.93 ±0.02 | 1,474 ±7 | 1.7% | 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 1,431 [1,428–1,434] Percentile 83 · #47 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
webdev/overall
|
1,519 ±3.34 | 1,455 ±1 | 46.0% | 2026-10-01 | xhigh-codex-harness |
| LMArena |
text/coding
|
1,483 ±3.46 | 1,456 ±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
B Agents 1,433 [1,428–1,438] Percentile 53 · #28 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
agent/overall
|
0.00 ±0.01 | 1,457 ±3 | 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 1,457 [1,450–1,465] Percentile 93 · #18 · components: 3
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,463 ±7.38 | 1,454 ±7 | 49.5% | 2026-10-02 | xhigh |
| Epoch AI |
frontiermath_t1_3
decay ×0.86 |
0.86 ±0.02 | 1,507 ±4 | 32.0% | 2026-07-09 | max |
| Epoch AI |
frontiermath_t4
decay ×0.86 |
0.71 ±0.07 | 1,505 ±12 | 11.0% | 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 1,435 [1,430–1,440] Percentile 71 · #33 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
vision/overall
|
1,271 ±3.87 | 1,449 ±4 | 63.4% | 2026-10-02 | xhigh |
| LMArena |
vision/ocr
|
1,278 ±4.12 | 1,450 ±4 | 15.6% | 2026-10-02 | xhigh |
| LMArena |
vision/diagram
|
1,297 ±6.15 | 1,460 ±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
B Writing 1,426 [1,421–1,431] Percentile 78 · #60 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,409 ±3.95 | 1,437 ±4 | 61.8% | 2026-10-02 | xhigh |
| LMArena |
text/longer_query
|
1,451 ±2.98 | 1,448 ±3 | 32.0% | 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 1,448 [1,440–1,457] Percentile 88 · #28 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,515 ±5.84 | 1,465 ±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 1,435 [1,416–1,454] Percentile 83 · #25 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,463 ±12 | 1,456 ±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
B Language: English 1,423 [1,418–1,429] Percentile 75 · #68 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,447 ±2.97 | 1,438 ±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 1,426 [1,410–1,442] Percentile 72 · #43 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,450 ±9.40 | 1,445 ±9 | 92.3% | 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
B Language: Français 1,434 [1,418–1,450] Percentile 77 · #32 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,473 ±9.65 | 1,454 ±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 1,440 [1,422–1,459] Percentile 84 · #22 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,447 ±14 | 1,462 ±10 | 91.6% | 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: 한국어 1,435 [1,419–1,451] Percentile 81 · #27 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,425 ±12 | 1,454 ±9 | 92.3% | 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
B Language: Polski 1,424 [1,405–1,444] Percentile 70 · #40 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,445 ±13 | 1,446 ±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: Русский 1,436 [1,428–1,444] Percentile 82 · #43 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,450 ±4.83 | 1,452 ±4 | 93.7% | 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
History
The same in a table
| Model | 2026-10 |
|---|---|
| GPT-5.6 Terra | 1,453 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.