Flux 1.1 Pro Ultra: a high-speed image generator now on Replicate
TL;DR
Black Forest Labs releases flux-1.1-pro-ultra on Replicate. FLUX.1.1 [pro] supports ultra and raw modes with images up to 4 megapixels.
What changed
Replicate launched black-forest-labs/flux-1.1-pro-ultra, featuring FLUX.1.1 [pro] in ultra and raw modes. This model generates images up to 4 megapixels. Vibe Builders gain direct access through Replicate's HTTP API or existing tokens.
Why it matters
Vibe Builders can now produce higher resolution images with improved realism in raw mode. The API integration fits existing stacks without setup changes. Developers benefit from quick deployment of advanced image generation.
What to watch for
Track Replicate's prediction costs for 4-megapixel outputs. Compare raw mode results against prior flux versions. Watch for community fine-tunes or mode-specific prompts.
Who this matters for
- Vibe Builders: Swap your existing Replicate model ID to flux-1.1-pro-ultra for higher resolution 4MP outputs.
- Basic Users: Use this model in Replicate's web interface to generate high-fidelity assets for marketing materials.
Harsh’s take
Black Forest Labs continues to set the pace for open-weight image generation. By pushing 4MP resolution directly into the Replicate ecosystem, they remove the friction of local hosting or complex infrastructure for high-quality visual assets. The raw mode is a direct shot at Midjourney, offering a more predictable, realistic output that businesses can actually use for professional design work.
However, the cost-per-generation remains the primary bottleneck for scaling. Developers must benchmark these ultra outputs against standard versions to ensure the increased resolution justifies the higher latency and price. Do not blindly switch your production pipelines to ultra mode until you verify that the performance overhead does not degrade your user experience. This is a powerful tool, but it requires disciplined cost management to remain viable.
by Harsh Desai
More AI news
- FeaturePerplexity releases Agent API to simplify multi-step AI agent workflows
Perplexity introduced the Agent API, a fully managed runtime designed to simplify the deployment and execution of complex multi-step AI agent workflows.
- FeatureLovable adds support for referencing exact code lines in chat prompts
Lovable now lets users point to specific lines of code by clicking the plus chip next to line numbers or using keyboard shortcuts. References appear as interactive pills in chat that jump to the code.
- FeatureCursor adds multi-select and continuous voice input to Design Mode
Cursor adds multi-select element support and continuous voice input to Design Mode in its browser. The updates let users queue changes without waiting for the agent to finish.