Skip to content
fedi.software

Mistral Small 3.2 in fedi-index

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

fedi-index · General

C 1,317 ±3

Percentile 38 · #183 · provisional

People prefer / Solves tasks

1,325 / —

Human votes vs benchmarks, Elo-eq

Value

-14

$0.15 per 1M tokens, median of 6 providers · vendor $0.15

On your hardware

Does not fit the reference setups.

Score decomposition

C General 1,317 [1,314–1,320] Percentile 38 · #183 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,338 ±2.65 1,338 ±3 26.5% 2026-10-02 —
LMArena text/hard_prompts 1,336 ±3.53 1,337 ±3 26.1% 2026-10-02 —
LMArena text/instruction_following 1,310 ±4.53 1,328 ±4 12.6% 2026-10-02 —
LMArena text/expert 1,293 ±10 1,307 ±8 10.5% 2026-10-02 —
LMArena text/longer_query 1,327 ±5.12 1,332 ±5 6.2% 2026-10-02 —
LMArena text/multi_turn 1,344 ±5.46 1,348 ±5 6.2% 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 Coding 1,333 [1,325–1,341] Percentile 37 · #175 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/coding 1,363 ±5.15 1,348 ±5 89.8% 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 1,318 [1,304–1,332] Percentile 37 · #161 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,341 ±8.95 1,337 ±9 85.8% 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 Vision 1,317 [1,311–1,323] Percentile 20 · #90 · components: 3 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena vision/overall 1,142 ±4.63 1,332 ±4 64.6% 2026-10-02 —
LMArena vision/ocr 1,134 ±5.70 1,311 ±6 15.4% 2026-10-02 —
LMArena vision/diagram 1,126 ±9.09 1,308 ±8 13.5% 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

C Writing 1,323 [1,314–1,331] Percentile 43 · #153 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,304 ±6.54 1,331 ±7 60.4% 2026-10-02 —
LMArena text/longer_query 1,327 ±5.12 1,332 ±5 32.8% 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 Language: 繁體中文 1,318 [1,306–1,332] Percentile 31 · #155 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,341 ±8.98 1,327 ±7 92.9% 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: Deutsch 1,337 [1,314–1,360] Percentile 30 · #101 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,341 ±15 1,352 ±13 90.2% 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: English 1,334 [1,327–1,341] Percentile 41 · #161 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,357 ±3.56 1,343 ±4 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

D Language: Español 1,334 [1,310–1,358] Percentile 16 · #126 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,347 ±15 1,349 ±14 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 Language: 日本語 1,323 [1,305–1,341] Percentile 24 · #97 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,276 ±13 1,334 ±10 91.7% 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: 한국어 1,317 [1,295–1,339] Percentile 21 · #108 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,260 ±16 1,329 ±12 90.5% 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: Polski 1,327 [1,316–1,339] Percentile 11 · #117 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,316 ±7.47 1,337 ±6 93.1% 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 Language: Русский 1,328 [1,315–1,342] Percentile 34 · #156 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,324 ±8.19 1,338 ±7 92.8% 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
Mistral Small 3.2 1,317 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: Mistral Small 3.2

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