Skip to contentT/TTokens or Towers

Compare

LeaderboardEvery tracked model, ranked and priced side by sideBenchmarksWhat each test measures and who leads itPricesAPI, media, and hardware rates with a cost calculator

Plan

Stack plannerSix questions to a priced local and hosted stackShortlistOur picks for subscriptions, APIs, and hardwareEnterpriseSeat plans against self-hosting for a whole team

Build

HardwareGPUs and workstations, and which models they fitHostsWhere to rent open-weight inference, and for how muchHarnessesCoding agents and the models they drive

Lab

FindingsResults from our own evaluationsScanAudit a repo’s AI usage in the browserMethodWhere the numbers come from and how we check them
Data checked September 25, 2026
Data checked September 25, 2026
Plan a stack
T/TTokens or Towers

An independent guide to AI models, what they cost, and the hardware to run them. No account, no API key, no sponsors.

Compare

  • Leaderboard
  • Benchmarks
  • Prices

Plan

  • Stack planner
  • Shortlist
  • Enterprise

Build

  • Hardware
  • Hosts
  • Harnesses

Lab

  • Findings
  • Scan
  • Method
Data checked September 25, 2026 · Every number links to its sourceRSS · Source on GitHub
Leaderboard/MiniMax
MiniMax

MiniMax · MiniMax H3

MiniMax H3

Open weightsGenerally availableNewMiniMax H3 Community License (custom)

MiniMax open-weights multimodal video model (33B-parameter H3-Omni-Transformer plus a Qwen3-VL-32B encoder) generating up to 15 s of 2K video with native stereo audio.

Best arena Elo
1357
Price
$0.13per second of 2K video (768p is $0.08/s)
Output speed
—
Output
video, audio

Image and video generation

  • Artificial Analysis Video Arena: text to video with audio#4 of 29
    1137independentwith audio; AA-Video-T2V v2.0 entry "MiniMax H3 (768p)"; rank 4; 95% CI ±9; 8315 appearancesArtificial Analysis Video Arena: Text to Video leaderboard ↗

    Leader: Wan 3.0 at 1156

  • Artificial Analysis Video Arena: text to video, silent#3 of 32
    1117independentwithout audio; AA-Video-T2V-Silent v2.0 entry "MiniMax H3 (768p)"; rank 3; 95% CI ±11; 4189 appearancesArtificial Analysis Video Arena: Text to Video leaderboard ↗

    Leader: Wan 3.0 at 1129

  • Artificial Analysis Video Arena: image to video with audio#2 of 26
    1181independentwith audio; AA entry "MiniMax H3"; rank 2; 95% CI ±8; 7080 appearancesArtificial Analysis Video Arena: Image to Video leaderboard ↗

    Leader: MiniMax H3 Max at 1194

  • Artificial Analysis Video Arena: image to video, silent#4 of 29
    1357independentwithout audio; AA entry "MiniMax H3"; rank 4; 95% CI ±11; 5679 appearancesArtificial Analysis Video Arena: Image to Video leaderboard ↗

    Leader: Gemini Omni 1.1 Flash at 1369

Specs

Released
July 31, 2026
Parameters
33B
Size class
10 to 40B
Input
text, image, audio, video
Output
video, audio
  • Price: MiniMax pay-as-you-go pricing ↗

Compare with

  • MiniMaxMiniMax H3 Max
  • GoogleGemini Omni 1.1 Flash
  • GoogleVeo 3.1
  • GoogleVeo 3.1 Fast