H3 Max Hub

What is H3 Max?

H3 Max is the speed-optimized build of MiniMax H3, the open-weight multimodal video model. It was jointly released by MiniMax and fal.ai on August 29, 2026. According to the source material, fal.ai post-trained MiniMax H3 for prompt adherence and visual quality and co-designed the inference stack for speed, enabling a 5-second 768p clip in roughly three seconds — faster than the video plays.

H3 Max is distinct from two related offerings. MiniMax H3 is the official open-weight model from MiniMax, announced July 31, 2026 (API), with weights published August 3 under the MiniMax H3 Community License. H3 Max Turbo is a further-accelerated variant served at minimax/h3-max-turbo, offering the lowest per-second price in the family.

How H3 Max works

H3 Max is available through fal.ai's hosted API only; no open weights are published for this variant. It supports text-to-video, image-to-video, and reference-to-video endpoints. Output resolutions are 480p and 768p. The model generates picture and audio in a single pass, so each clip carries its own soundtrack.

For comparison, MiniMax H3 supports text, images, video, and audio in one request and returns 5–15 seconds of video at 24 fps with native stereo audio. MiniMax H3 reaches up to 2K resolution via API, while its open checkpoints (FL2VA / Ref2VA) generate 768p natively. H3 Max is limited to 768p maximum.

Main features

• Speed: Approximately 3 seconds to generate a 5-second 768p clip.
• Output resolutions: 480p and 768p.
• Clip length: 5–15 seconds per request.
• Endpoints: Text-to-video, image-to-video, and reference-to-video.
• Native audio: Video and sound generated in a single pass.
• API only: No open weights for H3 Max.
• H3 Max Turbo: A further-accelerated variant with the same 480p/768p outputs and the lowest per-second price in the family.

Pricing

The supplied pages do not list specific prices for H3 Max or H3 Max Turbo. The source states that H3 Max Turbo offers the lowest per-second price in the family and advises checking fal.ai before budgeting, as pricing and limits change often. No pricing figures are available in the provided content.

Common use cases

• Fast production and interactive applications, where H3 Max's speed is beneficial.
• Real-time and interactive video streams, as noted for H3 Max.
• High-volume batch generation and live use, for which H3 Max Turbo is described as the best fit.
• Projects requiring native audio alongside video, since H3 Max generates both in one pass.

Limitations

• H3 Max is API only; no open weights are available for this variant.
• Maximum output resolution is 768p, lower than MiniMax H3's 2K API capability.
• Clip length is limited to 5–15 seconds per request.
• Pricing and limits are not specified in the provided pages and are subject to change; users should consult fal.ai directly.
• fal.ai has not published architectural details on how H3 Max Turbo differs from H3 Max.
• The h3max.info site is an independent community project, not affiliated with MiniMax or fal.ai.

Summary

H3 Max is a speed-focused variant of MiniMax H3, jointly released by MiniMax and fal.ai. It delivers 5-second 768p clips in about three seconds via hosted API endpoints for text-to-video, image-to-video, and reference-to-video. A faster H3 Max Turbo tier is available for volume and cost-sensitive use. Pricing details are not provided in the source material and should be verified with fal.ai.