Skip to content
fedi.software

Trinity Large Thinking in fedi-index

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

fedi-index · General

C 1,325 ±2

Percentile 40 · #176 · provisional

People prefer / Solves tasks

1,333 / —

Human votes vs benchmarks, Elo-eq

Value

-30

$0.3875 per 1M tokens, median of 5 providers · vendor $0.3875

On your hardware

  • 128 GB unified memory #41 · IQ2_XXS · strong compression · ≈31 tok/s

Score decomposition

C General 1,325 [1,323–1,328] Percentile 40 · #176 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,340 ±2.37 1,340 ±2 25.8% 2026-10-02 —
LMArena text/hard_prompts 1,341 ±2.85 1,342 ±3 25.7% 2026-10-02 —
LMArena text/instruction_following 1,324 ±3.49 1,341 ±3 12.6% 2026-10-02 —
LMArena text/expert 1,358 ±5.97 1,360 ±5 12.0% 2026-10-02 —
LMArena text/longer_query 1,318 ±3.41 1,324 ±3 6.3% 2026-10-02 —
LMArena text/multi_turn 1,320 ±4.55 1,326 ±4 6.1% 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

C Coding 1,344 [1,340–1,348] Percentile 41 · #164 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,238 ±10 1,375 ±3 45.4% 2026-10-01 —
LMArena text/coding 1,357 ±3.80 1,343 ±3 44.8% 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

C Math 1,340 [1,328–1,352] Percentile 43 · #145 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,366 ±7.58 1,361 ±7 86.6% 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

C Writing 1,322 [1,316–1,328] Percentile 43 · #154 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,307 ±4.79 1,333 ±5 61.2% 2026-10-02 —
LMArena text/longer_query 1,318 ±3.41 1,324 ±3 32.4% 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

C Language: 繁體中文 1,342 [1,332–1,353] Percentile 36 · #143 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,373 ±7.22 1,352 ±6 93.3% 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

C Language: Deutsch 1,349 [1,326–1,371] Percentile 34 · #96 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,356 ±15 1,364 ±13 90.3% 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

C Language: English 1,333 [1,327–1,339] Percentile 40 · #163 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,356 ±3.10 1,341 ±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

D Language: Español 1,321 [1,304–1,339] Percentile 14 · #129 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,328 ±11 1,332 ±10 91.8% 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

D Language: Français 1,342 [1,325–1,360] Percentile 19 · #109 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,365 ±10 1,354 ±10 91.9% 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

C Language: 한국어 1,350 [1,332–1,368] Percentile 27 · #100 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,306 ±13 1,364 ±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

D Language: Polski 1,321 [1,301–1,341] Percentile 9 · #119 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,312 ±13 1,332 ±11 91.0% 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

C Language: Русский 1,341 [1,332–1,351] Percentile 40 · #142 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,338 ±5.60 1,351 ±5 93.5% 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

fedi-index by month
The same in a table
Model 2026-10
Trinity Large Thinking 1,325 C

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: Trinity Large Thinking

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