Skip to content
fedi.software

MiniMax M2.7 in fedi-index

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

fedi-index · General

B 1,394 ±10

Percentile 64 · #105

People prefer / Solves tasks

1,402 / 1,375

Human votes vs benchmarks, Elo-eq

Value

+32

$0.525 per 1M tokens, median of 34 providers · vendor $0.525

On your hardware

  • GPU 12 GB #23 · UD-IQ1_M · strong compression · ≈16 tok/s
  • GPU 24 GB #25 · UD-IQ2_M · strong compression · ≈18 tok/s
  • 128 GB unified memory #3 · UD-IQ4_NL · ≈20 tok/s

Score decomposition

B General 1,394 [1,384–1,404] Percentile 64 · #105 · components: 7
Source Component Value Elo-eq Weight Measured Variant
Epoch AI eci 146 ±2.53 1,405 ±20 24.6% 2026-03-18 —
LMArena text/overall 1,405 ±1.73 1,405 ±2 19.2% 2026-10-02 —
LMArena text/hard_prompts 1,422 ±2.04 1,415 ±2 19.2% 2026-10-02 —
LMArena text/instruction_following 1,404 ±2.47 1,417 ±2 9.5% 2026-10-02 —
LMArena text/expert 1,443 ±3.79 1,430 ±3 9.4% 2026-10-02 —
LMArena text/longer_query 1,419 ±2.38 1,418 ±2 4.8% 2026-10-02 —
LMArena text/multi_turn 1,413 ±3.12 1,410 ±3 4.7% 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 Coding 1,403 [1,401–1,406] Percentile 69 · #86 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena webdev/overall 1,397 ±2.94 1,420 ±1 45.7% 2026-10-01 —
LMArena text/coding 1,453 ±2.60 1,429 ±2 44.8% 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 1,388 [1,379–1,396] Percentile 61 · #100 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/math 1,421 ±5.05 1,413 ±5 87.9% 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 Writing 1,382 [1,378–1,386] Percentile 57 · #115 · components: 2 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/creative_writing 1,354 ±3.19 1,382 ±3 62.1% 2026-10-02 —
LMArena text/longer_query 1,419 ±2.38 1,418 ±2 31.8% 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 Language: 繁體中文 1,393 [1,386–1,399] Percentile 53 · #105 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/chinese 1,439 ±4.59 1,405 ±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

C Language: Deutsch 1,384 [1,370–1,398] Percentile 48 · #75 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/german 1,395 ±8.82 1,398 ±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 1,402 [1,398–1,406] Percentile 64 · #100 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/english 1,425 ±2.28 1,415 ±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

C Language: Español 1,389 [1,378–1,400] Percentile 41 · #88 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/spanish 1,405 ±6.70 1,403 ±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

C Language: Français 1,393 [1,382–1,404] Percentile 41 · #80 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/french 1,421 ±6.59 1,406 ±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

D Language: 日本語 1,314 [1,299–1,330] Percentile 23 · #99 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/japanese 1,262 ±11 1,323 ±8 92.4% 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

C Language: 한국어 1,358 [1,345–1,370] Percentile 31 · #94 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/korean 1,313 ±8.87 1,369 ±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

C Language: Polski 1,385 [1,373–1,397] Percentile 35 · #86 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/polish 1,389 ±7.84 1,398 ±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

C Language: Русский 1,380 [1,374–1,386] Percentile 50 · #118 · components: 1 provisional
Source Component Value Elo-eq Weight Measured Variant
LMArena text/russian 1,383 ±3.51 1,392 ±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
MiniMax M2.7 1,394 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: MiniMax M2.7

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