Skip to content
fedi.software

mistral-large-3 in fedi-index

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

fedi-index · General

B 1,396 ±2

Percentile 66 · #101 · provisional

People prefer / Solves tasks

1,408 / —

Human votes vs benchmarks, Elo-eq

Value

+25

$0.75 per 1M tokens, median of 1 providers

On your hardware

Closed weights: cloud only.

Score decomposition

B General 1,396 [1,394–1,397] Percentile 66 · #101 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,428 ±1.46 1,428 ±2 25.6% 2026-10-02 —
LMArena text/hard_prompts 1,429 ±1.82 1,421 ±2 25.5% 2026-10-02 —
LMArena text/instruction_following 1,402 ±2.25 1,415 ±2 12.6% 2026-10-02 —
LMArena text/expert 1,421 ±3.71 1,412 ±3 12.4% 2026-10-02 —
LMArena text/longer_query 1,413 ±2.21 1,413 ±2 6.3% 2026-10-02 —
LMArena text/multi_turn 1,429 ±2.86 1,425 ±3 6.3% 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

B Coding 1,366 [1,362–1,371] Percentile 51 · #129 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,230 ±13 1,373 ±4 57.5% 2026-10-06 —
LMArena text/coding outlier 1,447 ±2.46 1,423 ±2 29.8% 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 Math 1,383 [1,375–1,390] Percentile 59 · #105 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,415 ±4.60 1,407 ±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

C Vision 1,392 [1,387–1,398] Percentile 43 · #65 · components: 3 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena vision/overall 1,222 ±4.13 1,404 ±4 63.7% 2026-10-02 —
LMArena vision/ocr 1,231 ±4.53 1,404 ±4 15.6% 2026-10-02 —
LMArena vision/diagram 1,241 ±7.04 1,410 ±6 14.4% 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

B Writing 1,401 [1,397–1,404] Percentile 64 · #96 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,385 ±2.89 1,414 ±3 62.2% 2026-10-02 —
LMArena text/longer_query 1,413 ±2.21 1,413 ±2 31.7% 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

B Language: 繁體中文 1,399 [1,393–1,405] Percentile 56 · #98 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,448 ±4.09 1,412 ±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

B Language: Deutsch 1,418 [1,404–1,432] Percentile 67 · #48 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,438 ±8.86 1,435 ±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

B Language: English 1,416 [1,412–1,420] Percentile 71 · #80 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,440 ±2.02 1,430 ±2 94.0% 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

B Language: Español 1,418 [1,407–1,429] Percentile 64 · #54 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,438 ±6.64 1,434 ±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

B Language: Français 1,422 [1,410–1,434] Percentile 66 · #46 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,456 ±7.12 1,438 ±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

B Language: 日本語 1,406 [1,391–1,421] Percentile 58 · #55 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,394 ±11 1,422 ±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

B Language: 한국어 1,408 [1,396–1,420] Percentile 60 · #55 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,385 ±8.80 1,424 ±7 93.0% 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

B Language: Polski 1,426 [1,415–1,437] Percentile 72 · #38 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,441 ±7.10 1,442 ±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

B Language: Русский 1,405 [1,399–1,410] Percentile 63 · #86 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,412 ±3.25 1,418 ±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

History

fedi-index by month
The same in a table
Model 2026-10
mistral-large-3 1,396 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: mistral-large-3

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