Skip to content
fedi.software

MiniMax M2 in fedi-index

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

fedi-index · General

C 1,330 ±4

Percentile 42 · #171 · provisional

People prefer / Solves tasks

1,338 / —

Human votes vs benchmarks, Elo-eq

Value

-32

$0.525 per 1M tokens, median of 16 providers · vendor $0.525

On your hardware

  • GPU 12 GB #28 · UD-TQ1_0 · strong compression · ≈17 tok/s
  • GPU 24 GB #31 · UD-IQ2_XXS · strong compression · ≈17 tok/s
  • 128 GB unified memory #26 · Q3_K_M · ≈20 tok/s

Score decomposition

C General 1,330 [1,326–1,334] Percentile 42 · #171 · components: 5 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/overall 1,340 ±3.93 1,340 ±4 30.6% 2026-10-02 —
LMArena text/hard_prompts 1,357 ±5.00 1,356 ±4 30.0% 2026-10-02 —
LMArena text/instruction_following 1,329 ±6.58 1,346 ±6 14.0% 2026-10-02 —
LMArena text/longer_query 1,332 ±7.39 1,337 ±7 6.8% 2026-10-02 —
LMArena text/multi_turn 1,362 ±9.04 1,364 ±8 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

C Coding 1,357 [1,351–1,363] Percentile 47 · #147 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,297 ±5.57 1,392 ±2 48.8% 2026-10-01 —
LMArena text/coding 1,370 ±7.67 1,354 ±7 40.9% 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 Math 1,315 [1,290–1,340] Percentile 36 · #164 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,350 ±17 1,345 ±16 78.7% 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

C Writing 1,313 [1,300–1,326] Percentile 40 · #160 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,288 ±10 1,314 ±10 57.0% 2026-10-02 —
LMArena text/longer_query 1,332 ±7.39 1,337 ±7 35.0% 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 Language: 繁體中文 1,335 [1,315–1,355] Percentile 36 · #144 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,367 ±14 1,348 ±11 91.1% 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: English 1,339 [1,329–1,350] Percentile 43 · #156 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,363 ±5.53 1,349 ±6 93.3% 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

C Language: Русский 1,332 [1,311–1,353] Percentile 36 · #151 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,332 ±13 1,345 ±12 90.8% 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
MiniMax M2 1,330 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: MiniMax M2

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