MiniMax M3 in fedi-index
The score in every slice, what it is made of, value, local run and history.
fedi-index · General
B 1,413 ±8
Percentile 75 · #73
People prefer / Solves tasks
1,424 / 1,392
Human votes vs benchmarks, Elo-eq
Value
+50
$0.525 per 1M tokens, median of 42 providers · vendor $0.525
On your hardware
Does not fit the reference setups.
Score decomposition
B General 1,413 [1,405–1,421] Percentile 75 · #73 · components: 8
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| Epoch AI |
eci
|
147 ±1.75 | 1,414 ±14 | 29.2% | 2026-06-01 | — |
| LMArena |
text/overall
|
1,432 ±2.00 | 1,432 ±2 | 17.7% | 2026-10-02 | — |
| LMArena |
text/hard_prompts
|
1,447 ±2.34 | 1,437 ±2 | 17.6% | 2026-10-02 | — |
| LMArena |
text/instruction_following
|
1,432 ±2.77 | 1,444 ±3 | 8.7% | 2026-10-02 | — |
| LMArena |
text/expert
|
1,461 ±4.15 | 1,444 ±3 | 8.6% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,444 ±2.66 | 1,442 ±2 | 4.4% | 2026-10-02 | — |
| LMArena |
text/multi_turn
|
1,441 ±3.50 | 1,436 ±3 | 4.3% | 2026-10-02 | — |
| Epoch AI |
gpqa_diamond
decay ×0.26 |
0.91 ±0.02 | 1,463 ±10 | 1.7% | 2026-08-10 | — |
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 1,421 [1,418–1,423] Percentile 78 · #63 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
webdev/overall
|
1,482 ±2.90 | 1,444 ±1 | 45.8% | 2026-10-01 | — |
| LMArena |
text/coding
|
1,469 ±2.96 | 1,443 ±3 | 44.6% | 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 Agents 1,402 [1,398–1,405] Percentile 19 · #47 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
agent/overall
|
-0.07 ±0.00 | 1,423 ±2 | 90.4% | 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 Math 1,395 [1,385–1,405] Percentile 67 · #85 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/math
|
1,430 ±5.82 | 1,422 ±6 | 87.5% | 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 Vision 1,420 [1,416–1,424] Percentile 57 · #49 · components: 3 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
vision/overall
|
1,254 ±3.34 | 1,434 ±3 | 63.1% | 2026-10-02 | — |
| LMArena |
vision/ocr
|
1,262 ±3.41 | 1,435 ±3 | 15.6% | 2026-10-02 | — |
| LMArena |
vision/diagram
|
1,268 ±4.89 | 1,434 ±4 | 15.1% | 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 Writing 1,422 [1,417–1,426] Percentile 76 · #64 · components: 2 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/creative_writing
|
1,405 ±3.48 | 1,433 ±4 | 62.0% | 2026-10-02 | — |
| LMArena |
text/longer_query
|
1,444 ±2.66 | 1,442 ±2 | 31.9% | 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: 繁體中文 1,411 [1,403–1,419] Percentile 67 · #75 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/chinese
|
1,464 ±5.33 | 1,425 ±4 | 93.7% | 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: Deutsch 1,407 [1,392–1,423] Percentile 62 · #56 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/german
|
1,426 ±10 | 1,424 ±9 | 92.3% | 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 Language: English 1,418 [1,414–1,424] Percentile 73 · #75 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/english
|
1,442 ±2.60 | 1,432 ±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: Español 1,411 [1,398–1,424] Percentile 56 · #66 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/spanish
|
1,431 ±7.83 | 1,427 ±7 | 92.8% | 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: Français 1,413 [1,400–1,425] Percentile 54 · #63 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/french
|
1,446 ±7.60 | 1,429 ±7 | 92.9% | 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 Language: 日本語 1,395 [1,379–1,410] Percentile 49 · #66 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/japanese
|
1,378 ±11 | 1,410 ±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
B Language: 한국어 1,401 [1,387–1,414] Percentile 53 · #65 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/korean
|
1,375 ±9.87 | 1,416 ±8 | 92.8% | 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: Polski 1,408 [1,393–1,422] Percentile 51 · #65 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/polish
|
1,420 ±9.47 | 1,424 ±8 | 92.6% | 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 1,418 [1,412–1,425] Percentile 72 · #65 · components: 1 provisional
| Source | Component | Value | Elo-eq | Weight | Measured | Variant |
|---|---|---|---|---|---|---|
| LMArena |
text/russian
|
1,428 ±4.02 | 1,432 ±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
History
The same in a table
| Model | 2026-10 |
|---|---|
| MiniMax M3 | 1,413 B |
Badge
Show the model's fedi-index on your site or in a README. The badge updates itself; it links to this page.
Data CC BY 4.0: keep the link to fedi.software.