ReelStack
READ
Long-form on the scene, the tools, and the people building it.
Manila's Creator Scene Is Not What You Think
The underground is surfacing — and it looks nothing like Silicon Valley's playbook.
Luma Ray3.14 Cut Video Costs 3x While Google Integrated Veo 3.1 Into Gemini Omni Flash
Luma released Ray3.14 with native 1080p generation and faster speeds, while Google paired Veo 3.1 with Gemini Omni Flash for multimodal video workflows.
Luma Brings Natural Language Editing to Dream Machine as ComfyUI Integrates Gemini Omni
Luma AI rolled out text-based video modifications in Dream Machine while Google brought Gemini Omni 1.1 Flash node controls to ComfyUI.
PixVerse R1 and Hunyuan3D 3.0 Lead a Shift to Real-Time AI Previs
PixVerse R1 introduces real-time video generation for shot planning, while Hunyuan3D 3.0 advances voxel precision for virtual production workflows.
Wan 2.7, LTX-2.3, and Seedance 2.5 Reframe Generative Control
A wave of flagship model updates introduces native multi-image grids, 30-second continuous generations, and direct first-to-last-frame conditioning.
Claude Fable 5 and GPT-5.6 Sol Face Off in AI Music Video Performance Benchmarks
A head-to-head comparison between Anthropic and OpenAI models reveals a narrowing gap in creative direction and asset generation for high-end music video production.
Google Veo 3 Enters Professional Production Trials as ComfyUI Bridges to AI Agents
Google DeepMind partners with A24 to test Veo 3 in film workflows while ComfyUI launches an MCP server for agentic video generation.
ComfyUI Integrates HappyHorse 1.1 and Krea 2 as Local Video Workflows Expand
The integration of HappyHorse 1.1 and Krea 2 into ComfyUI marks a shift toward unified, local, and audio-native AI filmmaking pipelines.
Google Veo 3.1 and Higgsfield Soul ID Solve Character Consistency for Filmmakers
Google and Higgsfield both launched major updates this week targeting the hardest problem in AI video: keeping characters and backgrounds consistent across multiple shots.
Higgsfield Unifies Video Models as Google Veo 3 Speed Increases Fourfold
Higgsfield integrates Kling 3.0 and Seedance 2.0 for multi-model workflows while Google DeepMind launches DiffusionGemma to accelerate Veo 3 generation speeds.
Ideogram 4.0 Releases as 9.3B Open-Weight Model with JSON Layout Control
Ideogram 4.0 shifts the text-to-image landscape by offering open-weights and structured JSON prompting for precise composition, with immediate support in ComfyUI.
Krea 2 and ComfyUI Open Local Workflows While ByteDance Faces Global Friction
Krea 2 launches its public API and LoRA training as NVIDIA brings real-time 4K upscaling and optimized local model support to ComfyUI.
The Toolkit Is the Argument
What you choose to build with says everything about what you believe is possible.
From Layers to Leaderboard
How a Cebu-based motion designer turned a tutorial series into a full-time drop economy.