Image created with gemini-2.5-flash-image with claude-sonnet-4-5. Image prompt: Cinematic nighttime field under expansive starry sky with bold white sans-serif text reading OPEN SOURCE centered as movie title typography, distant silhouettes of multiple people scattered across horizon all looking upward, subtle converging footpaths in moonlit grass, deep navy sky with silver stars, widescreen composition, film grain texture, high contrast minimalist poster aesthetic.
FLUX Playground – Black Forest Labs https://playground.bfl.ai/image/generate
FLUX.2: Frontier Visual Intelligence | Black Forest Labs https://bfl.ai/blog/flux-2
Overview – Black Forest Labs https://docs.bfl.ai/flux_2/flux2_overview
China just passed the U.S. in open model downloads for the first time 👀 New data from Economies of Open Intelligence led by @huggingface policy team & community collaborators, presents some notable observations: ✨ Developer adoption In 2025, Chinese model developers saw https://x.com/AdinaYakup/status/1993648553445527996
You can now run FP8 reinforcement learning on consumer GPUs! Try DeepSeek-R1’s FP8 GRPO at home using only a 5GB GPU. Qwen3-1.7B fits in 5GB VRAM. We collabed with PyTorch to make FP8 RL inference 1.4× faster. Unsloth: 60% less VRAM, 12× longer context. https://x.com/UnslothAI/status/1993358367776186801
Review of Deep Seek OCR | 90/30 Club https://lukeatkins.me/90_30_Club/posts/deepseekocr/
What will the next-gen LLM architecture look like? This question keeps sparking debates — and Zhihu contributor & developer Yuxuan offers a sharp comparison between DeepSeek Sparse Attention (DSA) and Native Sparse Attention (NSA), plus a practical look at implementing DSA https://x.com/ZhihuFrontier/status/1993231992876421156
black-forest-labs/FLUX.2-dev · Hugging Face https://huggingface.co/black-forest-labs/FLUX.2-dev
🎨 Introducing FLUX.2 from Black Forest Labs, a production-grade image generation model with multi-reference consistency and brand-exact color matching. AI natives can now use FLUX.2 on Together AI and benefit from reliable inference for production-scale visual content creation. https://x.com/togethercompute/status/1993464539871060400
black-forest-labs/flux2: Official inference repo for FLUX.2 models https://github.com/black-forest-labs/flux2
Check out @bfl_ml’s deep dive on the rate-distortion-modelability trade-off in diffusion latent space design. I previously discussed this in my blog post on latents. Here’s amazing practical demonstration, taken directly from the development of FLUX.2! https://x.com/sedielem/status/1993392284180865312
Diffusers welcomes FLUX-2 https://huggingface.co/blog/flux-2
fal/FLUX.2-Tiny-AutoEncoder · Hugging Face https://huggingface.co/fal/FLUX.2-Tiny-AutoEncoder
Flux.1-dev has been the second most liked model on Hugging Face just after Deepseek R1 so super excited to see the release of Flux.2-dev by @bfl_ml today! Download the weights or try the model (thanks to @fal) on @huggingface: https://x.com/huggingface/status/1993416469531836527
FLUX.2 image generation just dropped in Synthesia. 💥 From infographics to realistic images to wild abstract art — it’s all one text prompt away. Available now on all plans. https://x.com/synthesiaIO/status/1993364380629475655
FLUX.2 is available on Replicate. Multi-reference editing, complex typography, stunning detail up to 4MP. Available in 3 variants: FLUX.2 [pro] (original FLUX.2): https://x.com/replicate/status/1993345980784427029
FLUX.2 is here – our most capable image generation & editing model to date. Multi-reference. 4MP. Production-ready. Open weights. Into the new. https://x.com/bfl_ml/status/1993345470945804563
FLUX.2 is out, and AI Toolkit has day zero support for both text to image and editing.”” / X https://x.com/ostrisai/status/1993348832944697581
FLUX.2 is out! 🚨 FLUX.2-dev bing state-of-the-art to open weight, with a chonky 32B parameter model, however, it can run on low-end cards with quantization and the new remote text encoder Read more on the 🧨 diffusers welcomes FLUX-2 blog https://x.com/multimodalart/status/1993349396482998691
FLUX.2 just landed in LTX. We’re proud to be official launch partners with @bfl_ml FLUX.2 is built for production with speed, precision, and photorealism. Here’s what makes it different 🧵 https://x.com/LTXStudio/status/1993354049367031956
FLUX.2 Pro image model is now available on Vercel AI Gateway – Vercel https://vercel.com/changelog/flux-2-pro-image-model-is-now-available-on-vercel-ai-gateway
FLUX.2 Pro is now available on Vercel AI Gateway. • Text-to-image model designed for richer visualization and photo-quality detail generation • Set model to 𝚋𝚏𝚕/𝚏𝚕𝚞𝚡-𝟸-𝚙𝚛𝚘 https://x.com/vercel_dev/status/1993348429653930260
How to Train a FLUX.2 LoRA with AI Toolkit. In this tutorial I show you how to train a style LoRA for FLUX.2 that can be used in both text to image and image editing tasks. AI Toolkit has day 0 support for training @bfl_ml FLUX.2 LoRAs. Links in 🧵 https://x.com/ostrisai/status/1993360008218825185
Introducing Unlimited FLUX.2 Pro Multiple reference characters, sharper realism, smarter controls, and higher resolution More info below 👇 https://x.com/freepik/status/1993346586446119032
Prompting Guide – FLUX.2 – Black Forest Labs https://docs.bfl.ai/guides/prompting_guide_flux2
We have just open sourced the ‘Tiny Autoencoder’ for FLUX.2 [dev] that lets us stream intermediary outputs when generating images. This means you won’t have to deal with pesky progress bars anymore. https://x.com/fal/status/1993669462550323652
Great to see more compact OCR models landing in open source lately — and HunyuanOCR is a standout: a strong, versatile 1B model with impressive real-world coverage. The vLLM community already provides Day-0 support, so you can try it right away: https://x.com/vllm_project/status/1993291230558716237
deepseek-ai/DeepSeek-Math-V2 · Hugging Face https://huggingface.co/deepseek-ai/DeepSeek-Math-V2
deepseek-ai/DeepSeek-Math-V2 https://github.com/deepseek-ai/DeepSeek-Math-V2/tree/main
Continuous batching from first principles https://huggingface.co/blog/continuous_batching
This model is called Z-image, Apache2.0 licensed. 6B size, Coming soon.🫡 https://x.com/bdsqlsz/status/1993545608179990544
Leave a Reply