Black Forest Labs on 26 June released FLUX.1 Kontext [dev], a 12-billion-parameter image-editing model whose full weights are openly available under a commercial licence. The company says the system matches or exceeds the output quality of proprietary editors such as Google’s Gemini Flash and ByteDance’s Bagel, falling short only of GPT-4o and BFL’s own closed models. Built on multimodal flow-matching techniques, FLUX.1 can preserve character identity, perform local edits and transfer visual styles without additional fine-tuning. Guided-distillation training and a lightweight architecture allow the software to run on consumer-grade GPUs, broadening access for independent developers and smaller studios. Platform support followed within days: Together AI began hosting the full weights, and integrations have since been rolled out on Hugging Face, Krea AI, Heurist, Fireworks AI and Glif. The quick uptake underscores growing demand for open-weight alternatives that let enterprises customise models and keep data in-house, a capability typically unavailable with closed-source services.
Flux Kontext [dev] is now live on Glif. Combined with LoRAs, it enables insanely powerful workflow automations. Here's how to get the most of it👇 https://t.co/gowo4SpEtk
🏥Introducing MedGemma, part 2, including: 🔥A 27B multimodal MedGemma 👀MedSigLIP, a lightweight image/text encoder for medical image retrieval/classification 📜A technical report with details Blog: https://t.co/UGgTZOGrOo Paper: https://t.co/LGhr9ai1Hl https://t.co/HhqJjNOEeC
FLUX.1 Kontext Max and Pro from @bfl_ml are now on Fireworks! Generate images with character consistency, local editing, and style references. Read more https://t.co/KhoZehAQaq https://t.co/2opIoEXtof