Skip to content
fedi.software

MiMo-V2-Flash in fedi-index

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

fedi-index · General

B 1,387 ±2

Percentile 61 · #116 · provisional

People prefer / Solves tasks

1,398 / —

Human votes vs benchmarks, Elo-eq

Value

+55◆

$0.15 per 1M tokens, median of 6 providers · vendor $0.175

On your hardware

Does not fit the reference setups.

Score decomposition

B General 1,387 [1,385–1,389] Percentile 61 · #116 · components: 6 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,411 ±1.75 1,411 ±2 25.7% 2026-10-02 non-thinking
LMArena text/hard_prompts 1,420 ±2.19 1,413 ±2 25.6% 2026-10-02 non-thinking
LMArena text/instruction_following 1,392 ±2.78 1,406 ±3 12.6% 2026-10-02 non-thinking
LMArena text/expert 1,425 ±4.83 1,415 ±4 12.2% 2026-10-02 non-thinking
LMArena text/longer_query 1,409 ±2.74 1,409 ±3 6.3% 2026-10-02 non-thinking
LMArena text/multi_turn 1,404 ±3.50 1,402 ±3 6.2% 2026-10-02 non-thinking

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 1,390 [1,388–1,393] Percentile 62 · #106 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,330 ±4.95 1,401 ±1 45.7% 2026-10-01 non-thinking
LMArena text/coding 1,443 ±2.99 1,420 ±3 44.7% 2026-10-02 non-thinking

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 1,366 [1,357–1,375] Percentile 52 · #123 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,395 ±5.56 1,389 ±5 87.7% 2026-10-02 non-thinking

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 1,393 [1,388–1,398] Percentile 62 · #102 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,375 ±3.73 1,403 ±4 61.9% 2026-10-02 non-thinking
LMArena text/longer_query 1,409 ±2.74 1,409 ±3 32.0% 2026-10-02 non-thinking

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 1,410 [1,402–1,417] Percentile 65 · #79 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,462 ±5.26 1,423 ±4 93.7% 2026-10-02 non-thinking

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 1,406 [1,402–1,411] Percentile 66 · #94 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,430 ±2.46 1,420 ±3 94.0% 2026-10-02 non-thinking

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: Español 1,402 [1,388–1,415] Percentile 47 · #80 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,420 ±8.08 1,417 ±8 92.8% 2026-10-02 non-thinking

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: Français 1,391 [1,366–1,416] Percentile 39 · #83 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,429 ±16 1,413 ±14 89.3% 2026-10-02 thinking

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: 日本語 1,355 [1,335–1,375] Percentile 32 · #87 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,324 ±15 1,370 ±11 91.1% 2026-10-02 non-thinking

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: 한국어 1,388 [1,373–1,404] Percentile 45 · #75 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,358 ±11 1,404 ±8 92.4% 2026-10-02 non-thinking

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: Polski 1,398 [1,385–1,411] Percentile 43 · #75 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,407 ±8.59 1,413 ±7 92.8% 2026-10-02 non-thinking

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 1,383 [1,377–1,390] Percentile 52 · #114 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,387 ±3.91 1,395 ±4 93.8% 2026-10-02 non-thinking

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
MiMo-V2-Flash 1,387 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: MiMo-V2-Flash

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