Decision support

MiniMax H3 vs Hailuo 02 Pro

Compare canonical objects directly. The useful dimensions change with the object type, so model and hardware comparisons stay separate.

MiniMax H3

MiniMax

Sources:models.dev

Benchmark results

Head to head on 2 benchmarks both models report

  • MiniMax H32leads
  • Hailuo 02 Pro0leads
  • Arena Text to Video

    Arena rating · Higher is better

    • MiniMax H3
      Best1456
      +257 ahead · Arena #11CI 1448 to 1464
    • Hailuo 02 Pro
      1199
      Arena #36CI 1186 to 1211
    Runs and sources
    MiniMax H3
    • default145695% CI 1448 to 1464 · 29,257 votes
    Hailuo 02 Pro
    • default119995% CI 1186 to 1211 · 9,399 votes
  • Arena Image to Video

    Arena rating · Higher is better

    • MiniMax H3
      Best1493
      +264 ahead · Arena #3CI 1488 to 1499
    • Hailuo 02 Pro
      1230
      Arena #37CI 1219 to 1240
    Runs and sources
    MiniMax H3
    • default149395% CI 1488 to 1499 · 121,481 votes
    Hailuo 02 Pro
    • default123095% CI 1219 to 1240 · 21,743 votes
  • Arena Video Edit

    Arena rating · Higher is better

    • MiniMax H3
      1378
      Arena #5CI 1369 to 1388
    • Hailuo 02 Pro
      Not reported
    Runs and sources
    MiniMax H3
    • default137895% CI 1369 to 1388 · 8,385 votes
    Hailuo 02 Pro

    Not reported

Each model's best published result is shown; "Not reported" means no matched evaluation, not a score of zero. Arena and DeepSWE ranks are the leaderboards' own; other ranks are among catalogue models with a score. Results from different harnesses or settings are not directly comparable.

Arena ratings: Arena (lmarena-ai/leaderboard-dataset), licensed CC BY 4.0. Ratings come from pairwise human votes and are only comparable within one leaderboard; the interval is Arena's 95% confidence interval. Model names are matched to the Vertical Tabs catalog.

Model specifications

Release date
July 31, 2026
June 18, 2025
Context window
—
—
Native context
—
—
Max output tokens
—
—
Knowledge cutoff
—
—
Open weights
Open weights
Closed
Parameter count
33.1B
—

API pricing · USD per 1M tokens

Input
—
—
Output
—
—
Cached input
—
—
Cache write
—
—

Capabilities and modalities

Input modalities
text, image, video, audio
text, image
Output modalities
video, audio
video
Tool calling
—
—
Reasoning option
—
—
Supported options
—
—
Capabilities
—
—

Architecture and availability

Architecture
—
—
Model families
—
—
Tokenizer
—
—
Instruction format
—
—
License
MiniMax H3 Community License Agreement
Proprietary
Providers
0
0
Provider endpoints
0
0
Ollama variants
0
0
Metadata source
models.dev
arena

A dash means the field is not reported. Provider-specific performance and pricing may differ.

To give you the best experience, we use cookies for personalization, insights, and performance. Learn more.