Skip to content
fedi.software

Mistral Large 4 in fedi-index

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

fedi-index · General

B 1,424 ±9

Percentile 78 · #65

People prefer / Solves tasks

1,405 / 1,450

Human votes vs benchmarks, Elo-eq

Value

+47

$1.09 per 1M tokens, median of 14 providers · vendor $1.05

On your hardware

Closed weights: cloud only.

Score decomposition

B General 1,424 [1,415–1,433] Percentile 78 · #65 · components: 7
Source Component Value Elo-eq Weight Measured Variant
Epoch AI eci 155 ±1.47 1,474 ±12 38.9% 2026-10-29 —
LMArena text/overall 1,411 ±4.32 1,411 ±4 16.4% 2026-11-11 —
LMArena text/hard_prompts 1,434 ±5.19 1,426 ±5 16.2% 2026-11-11 —
LMArena text/instruction_following 1,409 ±6.73 1,421 ±6 7.5% 2026-11-11 —
LMArena text/expert 1,446 ±12 1,432 ±10 6.2% 2026-11-11 —
LMArena text/longer_query 1,404 ±6.26 1,404 ±6 3.8% 2026-11-11 —
LMArena text/multi_turn 1,421 ±10 1,418 ±9 3.2% 2026-11-11 —

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

A Coding 1,429 [1,424–1,434] Percentile 80 · #58 · components: 3
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,566 ±7.17 1,468 ±2 36.0% 2026-11-15 —
LMArena text/coding 1,457 ±7.99 1,432 ±7 30.0% 2026-11-11 —
Terminal-Bench tbench/terminal-bench decay ×0.90 0.28 ±0.02 1,450 ±4 24.8% 2026-10-06 max

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 Agents 1,409 [1,403–1,416] Percentile 42 · #46 · components: 2
Source Component Value Elo-eq Weight Measured Variant
LMArena agent/overall -0.07 ±0.01 1,423 ±5 66.1% 2026-11-15 —
Terminal-Bench tbench/terminal-bench decay ×0.90 0.28 ±0.02 1,450 ±4 24.8% 2026-10-06 max

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 1,396 [1,371–1,420] Percentile 64 · #92 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,456 ±16 1,447 ±16 79.1% 2026-11-11 —

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

C Vision 1,410 [1,401–1,419] Percentile 48 · #58 · components: 3 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena vision/overall 1,244 ±7.17 1,424 ±6 65.8% 2026-11-09 —
LMArena vision/ocr 1,254 ±8.60 1,427 ±8 15.0% 2026-11-09 —
LMArena vision/diagram 1,272 ±14 1,439 ±12 11.9% 2026-11-09 —

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 1,369 [1,359–1,380] Percentile 53 · #125 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,343 ±8.32 1,370 ±8 58.9% 2026-11-11 —
LMArena text/longer_query 1,404 ±6.26 1,404 ±6 33.8% 2026-11-11 —

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 Language: 繁體中文 1,401 [1,382–1,420] Percentile 56 · #97 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,457 ±13 1,419 ±10 91.5% 2026-11-11 —

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 Language: English 1,398 [1,386–1,411] Percentile 62 · #104 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,424 ±6.57 1,413 ±7 93.0% 2026-11-11 —

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 Language: Русский 1,402 [1,387–1,418] Percentile 60 · #92 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,414 ±9.55 1,419 ±9 92.3% 2026-11-11 —

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

History

fedi-index by month
The same in a table
Model 2026-11
Mistral Large 4 1,424 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 4

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