Black Forest Labs rumored to launch Flux 3

// 2h agoMODEL RELEASE

Black Forest Labs rumored to launch Flux 3

Black Forest Labs is rumored to be launching Flux 3 today, introducing a multimodal model that expands beyond text-to-image generation. The new model reportedly supports image generation, video clips up to 20 seconds long, and enhanced reasoning capabilities, signaling a major leap forward for the platform.

// ANALYSIS

A unified multimodal model offering 20-second video generation alongside reasoning could establish Flux 3 as a powerful contender in next-generation AI media creation.

–Extends the Flux model lineage into multimodal video generation and reasoning.
–Supports high-quality video generation up to 20 seconds in length.
–High community anticipation surrounding the release from Black Forest Labs (@bfl_ai).

// TAGS

flux-3black-forest-labsaimultimodalimage-genvideo-genreasoning

DISCOVERED

2h ago

2026-07-23

PUBLISHED

3h ago

2026-07-23

RELEVANCE

8/ 10

AUTHOR

mark_k

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

OPEN SOURCE40m ago

Graph Engineering course offers blueprint for AI agents

A viral X thread highlights an open-source repository titled "Graph Engineering Course" on GitHub, created by codejunkie99, which has gained thousands of stars. The course provides a step-by-step practical blueprint designed to help AI engineers and developers master graph-based architectures, knowledge graphs, and agentic workflows.

LAUNCH47m ago

Synara launches open-source multi-agent AI workspace

Synara is a free and open-source desktop application designed to streamline software development by unifying and orchestrating multiple AI coding agents—such as Claude Code, Cursor, and Gemini CLI—within a single interface. Operating completely local-first across macOS, Windows, and Linux, Synara combines chats, terminals, browser previews, code diffs, and isolated Git worktrees while enabling parallel agent execution and mid-conversation model handoffs.

RESEARCH1h ago

SLAI T-Rex enables DeepSeek-V4 post-training on Ascend SuperPODs

SLAI T-Rex is an open training framework enabling full-parameter post-training of DeepSeek-V4 MoE models on Huawei Ascend SuperPOD clusters, achieving 34.22% MFU (a 2.93x speedup). Paired with a 10K solver-verified synthetic dataset, fine-tuned DeepSeek-V4-Flash achieved a 71.81% zero-shot Pass@1 score on complex Operations Research tasks, outperforming GPT-5.4-Mini.