Black Forest Labs' Flux.2 Flex launches on Replicate today
TL;DR
Black Forest Labs launches Flux.2 Flex on Replicate. The model enables maximum-quality image generation and editing with support for ten reference images.
What changed
Black Forest Labs dropped flux-2-flex on Replicate. This model delivers max-quality image generation and editing. It handles up to ten reference images for detailed control.
Why it matters
Vibe Builders gain direct access via Replicate's HTTP API. Integrate it with your existing token in the stack. It boosts image workflows in your builds.
What to watch for
Track prediction speeds and pricing on Replicate. Experiment with multi-reference edits. Look for user-shared prompts and optimizations.
Who this matters for
- Vibe Builders: Integrate high-fidelity image editing into your apps using the Replicate HTTP API.
- Developers: Use the ten-image reference support to build complex generative workflows with Flux 2 Flex.
What to watch next
Black Forest Labs continues to set the standard for open weights, and Flux 2 Flex is a direct challenge to proprietary image models. By enabling multi-reference input, they have moved past simple text-to-image gimmicks into actual utility for production workflows. If your app relies on consistent character or style generation, this model is now the baseline you must beat.
Replicate makes this trivial to deploy, removing the infrastructure headache that usually accompanies high-end models. Stop wasting time fine-tuning smaller, inferior models when you can pipe requests to this endpoint today. The barrier to entry for high-quality visual generation is effectively zero, so your competitive advantage now depends entirely on how well you implement these reference images into your user experience.
by Harsh Desai
More from general
- Model ReleaseFlux 1.1 Pro Ultra releases on Replicate
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.
- Model ReleaseSenseNova U1 8B MoT unified multimodal model releases on Replicate
lucataco releases sensenova-u1-8b-mot on Replicate. SenseNova U1 8B MoT enables unified multimodal tasks including native text-to-image generation.
- Model ReleaseReplicate hosts lucataco/gemma-4-31B-IT, Google's 31B VLM
Replicate publishes lucataco/gemma-4-31B-IT. Google's open-weight Gemma 4 31B Instruct VLM processes image and text inputs to generate text outputs.