Author HubMatthias Bastian
Black Forest Labs has released FLUX.1 context [dev], a 12-billion parameter AI image model, for free non-commercial use on HuggingFace. It runs on standard hardware and supports ComfyUI, HuggingFace Diffusers, and TensorRT. According to Black Forest Labs and Artificial Analysis, it performs better than models like Gemini-Flash Image in most benchmarks. Optimized versions are available for Nvidia's new Blackwell architecture. For commercial use, licenses start at $999 per month via Black Forest Labs' license portal.
![Bar charts show FLUX.1 Context [dev] with peak values for character preservation, editing accuracy, style transfer, and text input compared to other image AI models.](https://the-decoder.com/wp-content/uploads/2025/07/BFL_benchmark_results.png)
The ERGO Innovation Lab and ECODYNAMICS teamed up to analyze how insurance content shows up in AI-powered search. Their study looked at over 33,000 AI search results and 600 websites, focusing on which types of content large language models like ChatGPT tend to surface. The results show that LLMs favor content that is easy to read, well-structured, and trustworthy - all traits associated with classic SEO. Modular content, presented in a question-and-answer style, and well-linked internally is also more likely to show up in AI-generated answers, just as in classic SEO.
The study also looked at hallucination rates. ChatGPT had the highest rate, with just under ten percent of its responses containing inaccuracies, while you.com delivered much more reliable results. These findings apply specifically to insurance-related queries.
