Tag
Modular 26.6 release opens the Mojo compiler for external contributions, adds developer experience improvements, and expands MAX with audio generation, new model support, and faster performance.
ElevenLabs has added image and video generation to its MCP platform, expanding beyond audio to support multi-modal content creation.
YuE2 is a new AI music generation model using symbolic planning that achieves competitive performance on benchmarks like WildSongBench, outperforming or matching proprietary systems such as Suno v5.
StepAudio 3 Gen is a general-purpose audio generation model that unifies text-to-speech, voice design, sound effects, and music within a single discrete autoregressive framework using residual vector quantization tokens.
An independent audio researcher has trained a model called Foundation-1 that generates infinite one-shot sounds for music production and converts text prompts into playable synths, releasing the model along with documentation and inferencing tools.
theDAW is an AI-powered DJ application that runs locally on PCs, integrating models like Stable Audio 3 and Magenta RealTime 2 for audio generation, and is interoperable with music software such as Ableton and Reaper.
Runway has released GWM Worlds 2, an interactive world model with real-time control capabilities, enabling the generation of 720p/24fps video and 48kHz audio on the fly. It allows users to manipulate the environment via text commands and integrate agents for task execution.
VIBE is an AI system that generates background music for videos by aligning with text instructions, using continuous latent modeling and reinforcement learning for better synchronization with video mood, timing, and genre.
Cartesia's TTS model Sonic-3.6 is now generally available, having been rebuilt from the architecture up and receiving positive feedback from its preview phase.
StabilityAI announces two new beta tools for Stable Audio 3.0: a DAW plugin for direct integration and an enhanced web experience for editing and generation, both commercially safe.
The Diffusers team announced several open-weight AI models including MiniMax H3, LTX 2.5, Wan Animate 2, and MiniMax Music 3, highlighting advancements in multimodal video and audio generation.
An early-preview LoRA for MiniMax-H3 enables 4-step audio-video generation instead of ~20 steps, yielding roughly 5x faster sampling, though quality is still immature.
Third-party ComfyUI-compatible LoRA conversions for MiniMax-H3 Turbo 4-step audio-video generation, including further-trained checkpoint-500 variants and an example workflow.
An early-preview LoRA for MiniMax-H3 that enables joint video and synchronized audio generation in 4 sampling steps instead of ~20, offering roughly 5x faster sampling, with ComfyUI custom nodes and three bf16 checkpoints.
This paper introduces SwanTale, a unified multi-speaker speech and audio generation model supporting both zero-shot and instruct tasks, along with SwanData-Caption for data annotation and SwanVAE for high-quality multi-audio-modality generation.
OpenMMLab has open-sourced Amphion, an audio generation toolbox supporting TTS, singing voice conversion, sound effect generation, and more. It is completely free for commercial use and supports local deployment.
Black Forest Labs announces FLUX 3, a multimodal foundation model that jointly learns from images, videos, and audio, enabling unified generation and understanding across modalities with early access now available.
Black Forest Lab's Flux 3 is a new omni-modal AI model capable of generating and predicting images, video, audio, and actions.
BFL has introduced FLUX 3, a multi-modal AI model capable of generating images, videos, and audio.
Lightricks releases LTX-2.5, an open-weights world model for generating synchronized video and audio from text, image, and video inputs, with features like native multishot generation and a new diffusion video decoder.