Tier
Moderate
Headquarters
Shanghai, China
Founded
2021
Flagship
MiniMax
Models tracked
7
Artificial Analysis
Top intelligence
54.7
Artificial Analysis Index
Cheapest model
$0.00
blended /1M
Fastest output
238
tokens / sec
Flagship
MiniMax’s top-scoring model on the Artificial Analysis Intelligence Index.
Most intelligent model
MiniMax-M3
Released 2026-06-01
Intelligence
54.7
Coding
43.4
Math
—
In $/1M
$0.30
Out $/1M
$1.20
Speed t/s
46
Benchmark breakdown
Models & benchmarks
Every MiniMax model on file — intelligence, coding, price, and speed.
| Model | Released | Intel | Coding | Math | In $/1M | Out $/1M | Speed t/s | Latency s |
|---|---|---|---|---|---|---|---|---|
| MiniMax-M3 | 2026-06-01 | 54.7 | 43.4 | — | $0.30 | $1.20 | 46 | 1.99 |
| MiniMax-M2.7 | 2026-03-18 | 49.6 | 41.9 | — | $0.30 | $1.20 | 61 | 1.33 |
| MiniMax-M2.5 | 2026-02-12 | 41.9 | 37.4 | — | $0.30 | $1.20 | 238 | 3.33 |
| MiniMax-M2.1 | 2025-12-23 | 39.4 | 32.8 | 82.7 | $0.30 | $1.20 | 224 | 2.59 |
| MiniMax-M2 | 2025-10-26 | 36.1 | 29.2 | 78.3 | $0.30 | $1.20 | 126 | 1.08 |
| MiniMax M1 80k | 2025-06-17 | 24.4 | 14.5 | 61.0 | $0.55 | $2.20 | 0 | 0.00 |
| MiniMax M1 40k | 2025-06-17 | 20.9 | 14.1 | 13.7 | $0.00 | $0.00 | 0 | 0.00 |
Trajectory
MiniMax’s model releases over time, by intelligence.
House strengths
Where MiniMax’s lineup over-indexes, averaged across its models.
Capability profile
Best value
MiniMax-M3
Most intelligence per dollar — 54.7 index at $0.53/1M blended.
104 index points per $1
Media models
Image, video, and speech models, ranked by arena Elo.
| Model | Type | Released | Elo | Rank |
|---|---|---|---|---|
| Hailuo 02 0616 | image to video | 2025-06 | 1270 | #15 |
| Hailuo 02 Pro | image to video | 2025-06 | 1244 | #31 |
| Hailuo 2.3 | image to video | 2025-10 | 1242 | #32 |
| Hailuo 2.3 Fast | image to video | 2025-10 | 1239 | #34 |
| Hailuo 02 Standard | image to video | 2025-06 | 1235 | #36 |
| Hailuo 02 Fast | image to video | 2025-06 | 1195 | #47 |
| Hailuo 02 Standard | text to video | 2025-06 | 1181 | #35 |
| Hailuo 2.3 | text to video | 2025-10 | 1180 | #37 |
| Speech 2.8 HD | text to speech | — | 1175 | #9 |
| Hailuo 02 Pro | text to video | 2025-06 | 1161 | #41 |
| Speech 2.8 Turbo | text to speech | — | 1151 | #14 |
| Speech 2.6 Turbo | text to speech | — | 1142 | #16 |
Overview
A Chinese AI company building long-context text, audio, and video generation models, with both API and open releases.
Visit MiniMax