MiniMax Token Plan: All AI Models Under One API Key
MiniMax upgraded its Coding Plan to the Token Plan, giving developers flat-rate access to Hailuo video, speech, music, image, and text models under one API key.
Blog
Deep dives, tutorials, and analysis for AI-powered creators.
Mistral Voxtral TTS Ships Open-Weights Speech Model
Mistral AI released Voxtral TTS, a 4-billion-parameter open-weights text-to-speech model that matches or beats ElevenLabs on naturalness benchmarks. It supports nine languages and clones voices from just three seconds of reference audio.
MagiHuman 15B Ships Open-Source Human Video AI
Researchers released daVinci-MagiHuman, a 15-billion-parameter open-source model that generates realistic human video with synchronized speech in two seconds.
Instagram AI Transitions Turn Still Photos Into Video
Instagram new AI Transition feature generates animated in-between frames from still photos, creating seamless video Stories without filming.
Midjourney V8 Alpha Adds Relax Mode and 4x Faster SREF
Midjourney has enabled relax mode for V8 Alpha and rolled out a significantly faster style reference system. The update gives Standard, Pro, and Mega subscribers access to unlimited V8 generations through relax queues, alongside a new SREF pipeline that runs 4x faster.
WordPress.com Opens MCP to AI Content Agents
WordPress.com now lets AI agents draft posts, build pages, manage comments, and organize content through its expanded Model Context Protocol integration.
Flowtonik Ships AI-Powered DAW for Music Producers
Stems Labs has launched Flowtonik, a macOS digital audio workstation with built-in AI assistants for stem generation, mixing, arrangement, and natural language editing.
xAI Grok Batch API Adds Image and Video Generation
xAI expanded its Grok Batch API to support image generation at two cents per image and video generation at five cents per second, all at 50 percent off standard rates.
Adobe Firefly 5 Launches With Custom Model Training
Adobe opened Firefly Custom Models to public beta, letting creators train private AI models on their own images, plus Firefly Image Model 5 and 30-plus third-party integrations.