Skip to content
fedi.software

snowflake-arctic-instruct in fedi-index

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

fedi-index · General

D 1,109 ±3

Percentile 13 · #256 · provisional

People prefer / Solves tasks

1,103 / —

Human votes vs benchmarks, Elo-eq

Value

—

On your hardware

Closed weights: cloud only.

Score decomposition

D General 1,109 [1,106–1,112] Percentile 13 · #256 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,101 ±3.00 1,101 ±3 27.3% 2026-10-02 —
LMArena text/hard_prompts 1,072 ±4.64 1,098 ±4 26.4% 2026-10-02 —
LMArena text/instruction_following 1,066 ±4.30 1,097 ±4 13.2% 2026-10-02 —
LMArena text/expert 1,028 ±8.76 1,090 ±7 11.6% 2026-10-02 —
LMArena text/multi_turn 1,072 ±6.39 1,101 ±6 6.2% 2026-10-02 —
LMArena text/longer_query outlier 999 ±7.25 1,026 ±7 2.9% 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 Coding 1,115 [1,106–1,124] Percentile 11 · #248 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,089 ±5.70 1,104 ±5 89.6% 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 Math 1,125 [1,115–1,134] Percentile 12 · #224 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,109 ±5.75 1,114 ±6 87.6% 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 Writing 1,097 [1,088–1,106] Percentile 10 · #242 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,081 ±6.12 1,102 ±6 73.9% 2026-10-02 —
LMArena text/longer_query outlier 999 ±7.25 1,026 ±7 18.0% 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

D Language: 繁體中文 1,116 [1,105–1,126] Percentile 9 · #204 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,067 ±7.00 1,109 ±6 93.4% 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

D Language: Deutsch 1,127 [1,110–1,145] Percentile 8 · #133 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,074 ±11 1,121 ±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: English 1,101 [1,093–1,108] Percentile 12 · #239 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,122 ±3.80 1,094 ±4 93.7% 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 1,120 [1,095–1,145] Percentile 3 · #145 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,088 ±16 1,110 ±14 89.3% 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

D Language: Français 1,146 [1,121–1,171] Percentile 5 · #128 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,131 ±16 1,140 ±14 89.4% 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

D Language: 日本語 1,113 [1,096–1,130] Percentile 3 · #124 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 969 ±13 1,105 ±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

D Language: 한국어 1,096 [1,076–1,116] Percentile 2 · #133 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 939 ±15 1,085 ±11 91.1% 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

D Language: Русский 1,144 [1,133–1,156] Percentile 9 · #212 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,103 ±6.78 1,140 ±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

History

fedi-index by month
The same in a table
Model 2026-10
snowflake-arctic-instruct 1,109 D

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: snowflake-arctic-instruct

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