heatdrop.ai

A curated prompt cookbook for OpenAI's latest image model, covering portraits, UI mockups, game screenshots, and posters you can drop straight into the API.

VoxCPM2 generates speech directly from text using continuous diffusion, no discrete audio tokens required.

This Codex Skill turns Chinese articles into hand-drawn, slightly absurd 16:9 illustrations where a deadpan black blob does the conceptual heavy lifting.

A research family of ASR and TTS models built on the bet that voice should be processed as long-form narrative, not chopped into seconds-long shards.

Cosmos 3 tries to unify video generation, robot action prediction, and physical reasoning inside a single 16B–64B Mixture-of-Transformers architecture.

ViMax orchestrates director, screenwriter, and producer agents to generate multi-shot videos from raw ideas, novels, or scripts.

OpenAI's Whisper replaces the usual Rube Goldberg pipeline of speech-processing tools with a single Transformer trained to do it all.

MisoTTS brings Sesame-style conversational speech synthesis to local hardware, with a Llama backbone and a stubbornly English-only vocabulary.

OmniVoice Studio runs voice cloning, dubbing, and dictation locally on macOS, Windows, and Linux — no API keys, no cloud, no subscription.

A visual programming interface for image, video, 3D, and audio generation that treats model pipelines as composable graphs.

An Electron app that wraps 200+ generative models behind a single UI, with an unusual pitch: no guardrails, no cloud lock-in, and a split personality between local and remote inference.

Pixelle-Video wires LLMs, image/video generators, TTS engines, and ffmpeg into a single Streamlit app that spits out short-form videos from a topic string.

A Chinese speech toolkit that bundles ASR, diarization, emotion detection, and streaming into one MIT-licensed package.

Curated prompts and API patterns for OpenAI's GPT-Image-2, organized by real use case rather than vibe or aesthetic.

Outpainting is the appetizer; the main course is automated 2D game asset generation with seam-aware tooling that exports engine-ready packs.

Voicebox bundles seven TTS engines, Whisper dictation, and MCP agent hooks into a single Tauri app — all offline.

A research framework that uses a multimodal LLM to plan video edits semantically, then hands off to a diffusion transformer to actually draw the frames.

A self-hostable infinite canvas that wires AI image generation, reference editing, and chat into one collaborative workspace.

Handy is an offline, open-source dictation app that pastes your words into any text field—built to be extended, not monetized.

Voice-Pro bundles Whisper, F5-TTS, CosyVoice, and a dozen other tools into a single Gradio interface for creators who want ElevenLabs-like results without the API bills.