Event date · · MiniMax

Reportedly: MiniMax open-sources H3 Image, a 4K image model generating in as fast as 10 seconds

Reported by ifanr · not yet confirmed by the company or a second independent outlet. We update this page when it is.

REPORTED

MiniMax has open-sourced H3 Image, a 4K image generation model that can produce images in as fast as 10 seconds, ifanr reported. The model combines GPT-level effects with Midjourney-like aesthetics. Availability outside China is not specified in the report.

China context

Original name
稀宇科技
Outside China
Not stated in the sources yet
Claims
Company-reported; not yet independently evaluated
For builders
If the open-source release is confirmed, developers outside China could access and fine-tune a 4K image model without API fees, but license terms must be verified.
For investors
The reported open-sourcing may pressure closed image model providers on pricing, but the lack of official confirmation makes the competitive impact uncertain.

Translated from Chinese. Quotes and facts link to the original sources.

What happened

ifanr reported that MiniMax has open-sourced H3 Image, a 4K image generation model capable of producing images in as fast as 10 seconds. The report describes the model as combining GPT-level effects with Midjourney-like aesthetics. No details on licensing, weights availability, or API access were provided.

Technical significance

The reported 10-second generation time for 4K images suggests efficient inference, but no architectural details or benchmark comparisons are provided in the evidence.

Industry impact

If confirmed, MiniMax's open-sourcing of H3 Image would give developers outside China access to a high-resolution image model without API costs, potentially undercutting closed competitors. Verification needed: check MiniMax's official channels for model weights and license terms.

Decision value

For developers, an open-source 4K image model could reduce image generation costs and enable customization. For investors, it signals MiniMax's strategy to compete through open weights, but the report lacks financial or adoption data.

What to watch

Watch for MiniMax's official announcement or model card on Hugging Face to confirm the open-source release and license. Also monitor whether the model appears in independent benchmark evaluations.

Latest in Chinese AI

  1. Ant GroupReportedly: Ant Group's Robbyant Open-Sources LingBot-Touch Tactile Model for Robot Hands
  2. HygonReportedly: Hygon Launches 14XX Industrial-Control CPU With 10W TDP and a -40°C to 105°C Operating Range
  3. Deep RoboticsReportedly: Deep Robotics releases Jueying X50 quadruped robot with 50kg payload and complex-terrain mobility
  4. Reportedly: Apple Intelligence generative AI service completes filing for iPad and MacBook in China
  5. ByteDanceReportedly: ByteDance releases Seedream 5.5 image model, integrated into Doubao, Doubao Work, and Jimeng

All China AI Events

AIGC Newsletter

China AI, with sources and context.

Analysis of Chinese AI models, companies and policy, and what you can use outside China.