Currently listed through these providers:
Model details
FLUX1.1 [pro]
FLUX1.1 pro represents Black Forest Labs' push toward high-performance image synthesis, built around a rectified flow-matching diffusion process running through a Transformer-based architecture sometimes described as DiT-type hybrids. The model carries roughly 12 billion parameters arranged in parallel blocks, a design choice that supports both fidelity and throughput. Developed under the internal codename "blueberry," it was crafted to serve creators, developers, and enterprises seeking scalable generative technology. Its feature set includes adjustable moderation controls, and later extensions introduced Ultra mode for resolutions up to 4 megapixels and a Raw mode tuned for natural photographic realism.
The FLUX1.1 pro release built directly on the earlier FLUX.1 pro foundation, inheriting its core architecture while making substantial gains in inference efficiency and output quality. According to Black Forest Labs, the newer model delivers six times faster generation than its predecessor while simultaneously improving image quality, prompt adherence, and output diversity. The model also entered the Artificial Analysis image arena under its codename, where it surpassed all competing entries to claim the highest overall Elo score on the leaderboard. FLUX.1 schnell, a sibling variant, carries Apache-2.0 open weights, while the development-grade model remains open but non-commercial, with production use governed through API partnerships. This positioning makes FLUX1.1 pro particularly attractive for teams that need rapid, reliable image generation without sacrificing visual consistency across creative pipelines.
Quick Info
Powered by- Provider
- Vercel AI Gateway
- Model key
- bfl/flux-pro-1.1
- Release date
- Oct 2, 2024
- Last updated
- Oct 1, 2024
- Input modalities
- Output modalities
- Capabilities
Limits
- Output tokens
- 0 tokens
- Context window
- 512 tokens
Latest news about FLUX1.1 [pro]
No articles yet. Fetch the latest news to show it here.