Skip to content
fedi.software

Claude Opus 4.1 in fedi-index

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

fedi-index · General

B 1,400 ±6

Percentile 69 · #92

People prefer / Solves tasks

1,418 / 1,375

Human votes vs benchmarks, Elo-eq

Value

-61

$30 per 1M tokens, median of 20 providers

On your hardware

Closed weights: cloud only.

Score decomposition

B General 1,400 [1,394–1,407] Percentile 69 · #92 · components: 8
Source Component Value Elo-eq Weight Measured Variant
Epoch AI eci 144 ±1.11 1,391 ±9 36.5% 2025-08-05 —
LMArena text/overall 1,419 ±1.77 1,419 ±2 16.1% 2026-10-02 thinking-16k
LMArena text/hard_prompts 1,443 ±2.32 1,433 ±2 16.0% 2026-10-02 thinking-16k
LMArena text/instruction_following 1,436 ±2.93 1,447 ±3 7.9% 2026-10-02 thinking-16k
LMArena text/expert 1,439 ±6.05 1,426 ±5 7.4% 2026-10-02 thinking-16k
LMArena text/longer_query 1,455 ±3.10 1,452 ±3 3.9% 2026-10-02 thinking-16k
LMArena text/multi_turn 1,444 ±3.49 1,439 ±3 3.9% 2026-10-02 thinking-16k
Epoch AI gpqa_diamond decay ×0.26 0.77 ±0.03 1,399 ±14 1.2% 2025-08-05 16k

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 Coding 1,417 [1,414–1,421] Percentile 76 · #67 · components: 3
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,389 ±5.46 1,418 ±2 39.6% 2026-10-01 —
LMArena text/coding 1,479 ±3.29 1,452 ±3 38.6% 2026-10-02 thinking-16k
Epoch AI swe_bench_verified decay ×0.37 0.73 ±0.02 1,449 ±10 13.4% 2026-02-11 —

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 Math 1,382 [1,376–1,388] Percentile 58 · #107 · components: 3
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,431 ±5.37 1,422 ±5 49.1% 2026-10-02 thinking-16k
Epoch AI frontiermath_t1_3 decay ×0.86 0.13 ±0.02 1,348 ±4 29.2% 2026-06-11 32k
Epoch AI frontiermath_t4 decay ×0.86 0.02 ±0.02 1,396 ±4 14.8% 2026-06-11 32k

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 1,430 [1,424–1,435] Percentile 79 · #57 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,412 ±3.91 1,441 ±4 61.9% 2026-10-02 thinking-16k
LMArena text/longer_query 1,455 ±3.10 1,452 ±3 31.9% 2026-10-02 thinking-16k

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: 繁體中文 1,384 [1,378–1,390] Percentile 50 · #112 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,427 ±4.14 1,395 ±3 93.8% 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

B Language: Deutsch 1,397 [1,382–1,412] Percentile 55 · #65 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,413 ±9.57 1,413 ±8 92.5% 2026-10-02 thinking-16k

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 1,409 [1,404–1,414] Percentile 67 · #90 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,433 ±2.44 1,422 ±3 94.0% 2026-10-02 thinking-16k

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 1,427 [1,415–1,438] Percentile 74 · #40 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,448 ±6.82 1,443 ±6 93.2% 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: Français 1,398 [1,380–1,416] Percentile 46 · #73 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,432 ±11 1,416 ±10 91.8% 2026-10-02 thinking-16k

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: 日本語 1,394 [1,378–1,410] Percentile 47 · #68 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,378 ±12 1,410 ±9 92.3% 2026-10-02 thinking-16k

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: 한국어 1,405 [1,392–1,418] Percentile 56 · #60 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,381 ±9.54 1,421 ±7 92.8% 2026-10-02 thinking-16k

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: Polski 1,405 [1,396–1,413] Percentile 47 · #70 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,413 ±5.21 1,418 ±4 93.6% 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

B Language: Русский 1,412 [1,406–1,418] Percentile 68 · #75 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,421 ±3.64 1,426 ±3 93.8% 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

History

fedi-index by month
The same in a table
Model 2026-10
Claude Opus 4.1 1,400 B

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: Claude Opus 4.1

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