Skip to content
fedi.software

mixtral-8x22b-instruct-v0.1 in fedi-index

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

fedi-index · General

D 1,169 ±3

Percentile 20 · #234 · provisional

People prefer / Solves tasks

1,167 / —

Human votes vs benchmarks, Elo-eq

Value

-236

$3 per 1M tokens, median of 1 providers

On your hardware

Does not fit the reference setups.

Score decomposition

D General 1,169 [1,167–1,172] Percentile 20 · #234 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,162 ±2.31 1,162 ±2 26.1% 2026-10-02 —
LMArena text/hard_prompts 1,150 ±3.53 1,169 ±3 25.5% 2026-10-02 —
LMArena text/instruction_following 1,147 ±3.30 1,174 ±3 12.8% 2026-10-02 —
LMArena text/expert 1,113 ±6.98 1,160 ±6 11.7% 2026-10-02 —
LMArena text/multi_turn 1,130 ±4.58 1,154 ±4 6.2% 2026-10-02 —
LMArena text/longer_query 1,144 ±4.97 1,161 ±5 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

D Coding 1,175 [1,168–1,182] Percentile 19 · #225 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,166 ±4.36 1,172 ±4 90.0% 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 Math 1,188 [1,181–1,196] Percentile 22 · #198 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,184 ±4.57 1,186 ±4 88.0% 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 Writing 1,166 [1,159–1,172] Percentile 22 · #210 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,141 ±4.72 1,164 ±5 62.2% 2026-10-02 —
LMArena text/longer_query 1,144 ±4.97 1,161 ±5 31.3% 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: 繁體中文 1,152 [1,144–1,160] Percentile 12 · #197 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,116 ±5.68 1,148 ±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

D Language: Deutsch 1,181 [1,167–1,194] Percentile 13 · #125 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,141 ±8.82 1,179 ±8 92.7% 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: English 1,166 [1,160–1,171] Percentile 22 · #214 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,187 ±2.88 1,163 ±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 1,171 [1,150–1,192] Percentile 7 · #139 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,150 ±13 1,168 ±12 90.7% 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: Français 1,174 [1,152–1,196] Percentile 9 · #123 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,165 ±13 1,171 ±12 90.6% 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: 日本語 1,159 [1,144–1,174] Percentile 7 · #119 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,037 ±11 1,155 ±8 92.5% 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

D Language: 한국어 1,176 [1,161–1,192] Percentile 9 · #124 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,056 ±11 1,174 ±8 92.4% 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: Русский 1,190 [1,181–1,199] Percentile 13 · #204 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,158 ±5.21 1,189 ±5 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

History

fedi-index by month
The same in a table
Model 2026-10
mixtral-8x22b-instruct-v0.1 1,169 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: mixtral-8x22b-instruct-v0.1

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