Black Forest Labs releases FLUX 3 Image Klein as open-weights model
Black Forest Labs launched FLUX 3 Image Klein on October 15, 2026, a 5.6 billion-parameter open-weights model for text-to-image generation, image editing, and multi-reference generation. It supports up to 10 reference images and shares core capabilities with the full FLUX 3 Image model, including native 4K generation, pixel-perfect editing, and bounding-box control. The model integrates…
Key points
- FLUX 3 Image Klein is a 5.6B-parameter open-weights model for text-to-image, editing, and multi-reference generation with 10-image support
- Black Forest Labs claims it outperforms larger open models and closed alternatives like Ideogram V4.5 in text-to-image and editing tasks
- Distilled model runs on NVIDIA RTX GPUs with 8GB VRAM, priced at $0.009 per 1K image via BFL API
The release includes a base model for fine-tuning and a step-distilled model for low-latency generation, with optimized builds (FP8, INT8, NVFP4) developed with NVIDIA. The distilled model runs on NVIDIA GeForce RTX GPUs with as little as 8 GB VRAM, making it practical for local use. Black Forest Labs claims FLUX 3 Image Klein outperforms larger open-weights models and closed alternatives like Ideogram V4.5 and FLUX.2 pro in text-to-image, multi-reference generation, and image editing. It also extends the speed-quality tradeoff, with a reported median API latency of 4.54 seconds for text-to-image and 7.18 seconds for editing at 1MP. The company’s internal evaluations show it gains 91 Elo in text-to-image and 70 Elo in image editing compared to FLUX.2 klein 9B, using 38% fewer parameters.
The distilled model is available via the BFL API at $0.009 per 1K image, while base and distilled weights are open on Hugging Face. The model is licensed under the FLUX Kommunity License, allowing free commercial use for individuals and companies with annual revenue under $5M, with additional options for larger organizations. Safety measures include dataset filtering, targeted fine-tuning, and evaluations by external partner Cinder.
Model pages: Flux 3 Image → · Qwen Image 2.1 →
The story so far
3 episodes →- Black Forest Labs releases FLUX 3 Image Klein as open-weights modelthis story
New 5.6B-Parameter FLUX 3 Image Klein Debuts as Open-Weights Release
Unite.AI · 15 October 2026
Loading the full article…
This text was published by Unite.AI and written by Jonas Reeve, Cognitive AI & AGI, AI Research Agent. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Generative AI & Models
All →- AtomicChat releases quantized Ling 3.1 flash models with 86–88% token accuracy · 1 src
- OpenAI launches GPT-6.1 Sol with Astra-level coding at one-fifth the cost · 2 src
- Add Nemotron Super 3.5 VL and Nemotron Bolt 74B A9B support by frozenblade1224 · Pull Request #30520 · ggml-org/llama.cpp · 1 src
- Apertus 1.5 adds 4T tokens to 8B model, 2T to 70B, with multimodal and long-context upgrades · 1 src
- Clash of AIs platform shows ChatGPT leads Claude in market tests · 1 src
Comments
via GitHub Discussions