Categories
Categories
Creative lanes first: video, image, speech, music, games.
Open-source image generation UIs, pipelines and models, from node-based workflows to one-click web UIs. 11 overall
SmolVLM2-2.2B-Instruct 11
HuggingFaceTB/SmolVLM2-2.2B-Instruct A vision-language model that fits on a laptop and reads video. MiniCPM-V-2_6 10
openbmb/MiniCPM-V-2_6 A 3B model that sees, hears and reads — with excellent OCR. dicebear 80
dicebear/dicebear Deterministic avatars and illustrations for developers — 30+ styles, one API. image-to-3dlab 78
Bingeljell/image-to-3dlab Drop an image, get a game-ready 3D asset — on Apple Silicon. Real-ESRGAN 16
xinntao/Real-ESRGAN Practical super-resolution for photos, anime and frames InstantID 10
instantX-research/InstantID Keep the same face across generated images without heavy fine-tuning TRELLIS 58
microsoft/TRELLIS Structured image-to-3D for game props TripoSR 58
VAST-AI-Research/TripoSR Fast feed-forward image-to-3D InstantMesh 13
TencentARC/InstantMesh Efficient image-to-3D meshes stable-dreamfusion 7
ashawkey/stable-dreamfusion Classic open text-to-3D playground Omost 8
lllyasviel/Omost LLM-composed canvases for structured concept art Open video generation models and tools. Hardware requirements vary a lot, so check the setup card first. 9 overall
ArcReel 82
ArcReel/ArcReel Turn a novel or screenplay into a finished video, on your own server. dramaclaw 81
dramaclaw/dramaclaw A general-purpose AIGC video engine: script to finished film in one pipeline. FunClip 72
modelscope/FunClip Transcribe, subtitle and LLM-clip long videos — locally. auto-editor 79
WyattBlue/auto-editor Effort-free video editing: it cuts the dead air for you. reelmimic 60
edenfunf/reelmimic Show it a video you love; get a new video in the same style. whisperX 75
m-bain/whisperX Word-level timestamps on top of Whisper-class ASR SadTalker 14
OpenTalker/SadTalker Talking-head video from one photo plus audio MotionCtrl 8
TencentARC/MotionCtrl Controllable camera/object motion for AI video i2vgen-xl 7
ali-vilab/i2vgen-xl Image-to-video for animating concept art Speech recognition and text-to-speech projects. Language coverage and license terms matter as much as quality here. 9 overall
Kokoro-82M 19
hexgrad/Kokoro-82M Studio-quality TTS from an 82M model you can run on a CPU. DiffSinger 63
openvpi/DiffSinger Diffusion-based singing voice synthesis toolkit F5-TTS 70
SWivid/F5-TTS Fast zero-shot TTS with natural prosody fish-speech 75
fishaudio/fish-speech Multilingual open TTS with an active community OpenVoice 20
myshell-ai/OpenVoice Instant voice cloning with style control Retrieval-based-Voice-Conversion-WebUI 65
RVC-Project/Retrieval-based-Voice-Conversion-WebUI The default WebUI for voice conversion covers whisperX 75
m-bain/whisperX Word-level timestamps on top of Whisper-class ASR voice-changer 60
w-okada/voice-changer Real-time AI voice changer for streams and games so-vits-svc 13
svc-develop-team/so-vits-svc Classic open singing voice conversion stack Compose, accompany, sample and transform music with open models and tools — separate from speech TTS/ASR. 17 overall
ACE-Step-1.5 73
ace-step/ACE-Step-1.5 Local full-song music generation that fits on a consumer GPU YuE 78
multimodal-art-projection/YuE Open music model with symbolic planning and zero-shot covers ACE-Step 35
ace-step/ACE-Step Earlier ACE-Step stack — still useful if you follow v1 tutorials demucs 19
facebookresearch/demucs Split any track into vocals, drums, bass and other stems stable-audio-tools 60
Stability-AI/stable-audio-tools Training and inference toolkit for Stable Audio open models audiocraft 42
facebookresearch/audiocraft MusicGen and companion models for text-to-music DiffSinger 63
openvpi/DiffSinger Diffusion-based singing voice synthesis toolkit ACE-Step-Studio 75
timoncool/ACE-Step-Studio One-click Windows studio built on ACE-Step 1.5 XL Retrieval-based-Voice-Conversion-WebUI 65
RVC-Project/Retrieval-based-Voice-Conversion-WebUI The default WebUI for voice conversion covers DiffRhythm2 28
ASLP-lab/DiffRhythm2 Full songs with lyric alignment via block flow matching heartlib 62
HeartMuLa/heartlib Open lyric+tag music generation stack for 2026 Muse 31
yuhui1038/Muse Fully open long-form song model with style controls AudioSep 7
Audio-AGI/AudioSep Text-queried universal audio separation AudioLDM2 8
haoheliu/AudioLDM2 Latent diffusion for text-to-audio and music ultimatevocalremovergui 66
Anjok07/ultimatevocalremovergui Desktop GUI for high-quality vocal/stem removal voice-changer 60
w-okada/voice-changer Real-time AI voice changer for streams and games so-vits-svc 13
svc-develop-team/so-vits-svc Classic open singing voice conversion stack Tools for making and restyling pictures by hand or by code: pixel art editors, procedural avatars and style-mimicry pipelines. 12 overall
reelmimic 60
edenfunf/reelmimic Show it a video you love; get a new video in the same style. dicebear 80
dicebear/dicebear Deterministic avatars and illustrations for developers — 30+ styles, one API. Pixelorama 81
Orama-Interactive/Pixelorama A free, open pixel art multitool — sprites, animation and palettes. PixiEditor 77
PixiEditor/PixiEditor One universal 2D editor: pixels, vectors, animation and node graphs. SadTalker 14
OpenTalker/SadTalker Talking-head video from one photo plus audio Real-ESRGAN 16
xinntao/Real-ESRGAN Practical super-resolution for photos, anime and frames InstantID 10
instantX-research/InstantID Keep the same face across generated images without heavy fine-tuning texel-studio 59
EYamanS/texel-studio Pixel-art agent that paints sprites tool-by-tool spritebrew 67
GAlbanese09/spritebrew Generate and export game-ready pixel sprite sheets spritesage 68
keystoneintelligence/spritesage Sprite editor with AI assist and Godot export Omost 8
lllyasviel/Omost LLM-composed canvases for structured concept art PuLID 3
ToTheBeginning/PuLID PuLID: high-fidelity identity customization for diffusion — consistent characters for games and comics. Build games faster with AI: asset generation, image-to-3D pipelines, AI modders and one-person studio toolkits. 14 overall
universal-modder 79
rehan-remade/universal-modder Point Claude at any game — and let it mod almost anything. skills 78
scenario-labs/skills Production-ready images, video, audio and 3D — generated from any agent. image-to-3dlab 78
Bingeljell/image-to-3dlab Drop an image, get a game-ready 3D asset — on Apple Silicon. ohmygame 76
WhiteTowerAI/ohmygame Your own one-person game studio, orchestrated by AI. texel-studio 59
EYamanS/texel-studio Pixel-art agent that paints sprites tool-by-tool spritebrew 67
GAlbanese09/spritebrew Generate and export game-ready pixel sprite sheets spritesage 68
keystoneintelligence/spritesage Sprite editor with AI assist and Godot export TRELLIS 58
microsoft/TRELLIS Structured image-to-3D for game props TripoSR 58
VAST-AI-Research/TripoSR Fast feed-forward image-to-3D InstantMesh 13
TencentARC/InstantMesh Efficient image-to-3D meshes stable-dreamfusion 7
ashawkey/stable-dreamfusion Classic open text-to-3D playground PuLID 3
ToTheBeginning/PuLID PuLID: high-fidelity identity customization for diffusion — consistent characters for games and comics. MotionCtrl 8
TencentARC/MotionCtrl Controllable camera/object motion for AI video i2vgen-xl 7
ali-vilab/i2vgen-xl Image-to-video for animating concept art Open-source large language models, runtimes and serving engines. Compare them by speed, hardware requirements and how actively they are maintained. 5 overall
lmdeploy 71
InternLM/lmdeploy Turbocharge LLM/VLM serving with 4-bit weights and a fast gRPC API. SmolLM2-1.7B-Instruct 0
HuggingFaceTB/SmolLM2-1.7B-Instruct A 1.7B model good enough to be your default on-device LLM. Phi-3.5-mini-instruct 15
microsoft/Phi-3.5-mini-instruct A 3.8B model that reasons better than models twice its size. Qwen2.5-3B-Instruct 0
Qwen/Qwen2.5-3B-Instruct A 3B all-rounder with unexpectedly good coding and 29+ languages. gemma-3-1b-it 13
google/gemma-3-1b-it A 1B model that reads images and speaks 140+ languages with a 128K context. Frameworks for building AI agents that plan, call tools and collaborate. Look at abstraction level, multi-agent support and production readiness before you pick one. 8 overall
screenpipe 73
screenpipe/screenpipe Record everything on your screen locally and let AI search and act on it. gpt-researcher 72
assafelovic/gpt-researcher Point it at a question; it browses the web and hands back a sourced report. e2b 78
e2b-dev/e2b Give your agent a disposable Linux sandbox to run untrusted code safely. khoj 56
khoj-ai/khoj Ask questions across your own notes, files and chats from a chat box. litellm 80
BerriAI/litellm One import to talk to OpenAI, Anthropic, Google, Mistral and 100+ more. skills 62
anthropics/skills The canonical Skill library — copy any SKILL.md and your agent gains a capability. superpowers 69
obra/superpowers Drop-in skills that make your coding agent actually do TDD and debug properly. planning-with-files 68
OthmanAdi/planning-with-files Give your agent a durable, file-backed plan it actually follows. Retrieval-augmented generation toolkits and ready-to-run knowledge-base apps. The key differences are document parsing quality, retrieval strategy and deployment effort. 6 overall
screenpipe 73
screenpipe/screenpipe Record everything on your screen locally and let AI search and act on it. gpt-researcher 72
assafelovic/gpt-researcher Point it at a question; it browses the web and hands back a sourced report. WrenAI 77
Canner/WrenAI Ask your database questions in plain English; it clarifies, then writes the SQL. docling 78
docling-project/docling Turn messy PDFs and Office docs into clean Markdown, JSON or a layout tree. khoj 56
khoj-ai/khoj Ask questions across your own notes, files and chats from a chat box. txtai 72
neuml/txtai Embed, index, search and prompt — no separate vector DB to run. Reusable skills, prompts and plugins that extend what coding and general-purpose agents can do. 10 overall
skills 62
anthropics/skills The canonical Skill library — copy any SKILL.md and your agent gains a capability. superpowers 69
obra/superpowers Drop-in skills that make your coding agent actually do TDD and debug properly. Skill_Seekers 63
yusufkaraaslan/Skill_Seekers Point it at a docs site and get a ready-to-use SKILL.md back. ui-ux-pro-max-skill 71
nextlevelbuilder/ui-ux-pro-max-skill Make your agent stop shipping generic 'AI slop' front-ends. planning-with-files 68
OthmanAdi/planning-with-files Give your agent a durable, file-backed plan it actually follows. scientific-agent-skills 73
K-Dense-AI/scientific-agent-skills Domain skills that let an agent do real scientific analysis. humanizer 70
blader/humanizer Make AI drafts read like a human wrote them. awesome-claude-skills 46
travisvn/awesome-claude-skills The discovery index for finding your next agent skill. universal-modder 79
rehan-remade/universal-modder Point Claude at any game — and let it mod almost anything. skills 78
scenario-labs/skills Production-ready images, video, audio and 3D — generated from any agent. Open-source coding agents and editor extensions. Compare model flexibility, autonomy level and cost control. 3 overall
e2b 78
e2b-dev/e2b Give your agent a disposable Linux sandbox to run untrusted code safely. superpowers 69
obra/superpowers Drop-in skills that make your coding agent actually do TDD and debug properly. ui-ux-pro-max-skill 71
nextlevelbuilder/ui-ux-pro-max-skill Make your agent stop shipping generic 'AI slop' front-ends. Run models on your own laptop, workstation or server. We highlight minimum VRAM, platform support and how fast you can get a first answer. 1 overall
Models and tools that work across text, images, audio and video. 3 overall
SmolVLM2-2.2B-Instruct 11
HuggingFaceTB/SmolVLM2-2.2B-Instruct A vision-language model that fits on a laptop and reads video. gemma-3-1b-it 13
google/gemma-3-1b-it A 1B model that reads images and speaks 140+ languages with a 128K context. MiniCPM-V-2_6 10
openbmb/MiniCPM-V-2_6 A 3B model that sees, hears and reads — with excellent OCR. Datasets, benchmarks and evaluation harnesses for measuring model quality. 2 overall
Self-hostable platforms for building chat apps, agents and automations without writing everything from scratch. 5 overall
WrenAI 77
Canner/WrenAI Ask your database questions in plain English; it clarifies, then writes the SQL. ArcReel 82
ArcReel/ArcReel Turn a novel or screenplay into a finished video, on your own server. dramaclaw 81
dramaclaw/dramaclaw A general-purpose AIGC video engine: script to finished film in one pipeline. ohmygame 76
WhiteTowerAI/ohmygame Your own one-person game studio, orchestrated by AI. ACE-Step-Studio 75
timoncool/ACE-Step-Studio One-click Windows studio built on ACE-Step 1.5 XL Vector databases and similarity search engines that power RAG and recommendation systems. 2 overall
Agents that operate browsers and desktops the way people do: clicking, typing and reading the screen. 1 overall