ComfyUI
The nodal workflow engine for serious diffusion.
Open Source 6–16 GB VRAM
- Min VRAM
- 6 GB
- GPU class
- Mid GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
For complex AI workflows & local compute
The right tool for your VRAM budget.
A curated guide to ComfyUI workflows, local LLM runners, and heavy image/video diffusion — with hardware requirements front and centre. No fake free trials. No “as a service” sludge.
18+ tools tracked11 categoriesHardware-first verificationNo paid placements
What do you want to do?
All categories →Generate 3DGenerate meshes, NeRFs, Gaussian splats, and avatars from images or text.Build AI agentsBuild autonomous AI agents that reason, plan, and use tools.Edit & upscale videoUpscaling, inpainting, restyle, frame interpolation.Generate audioText-to-speech, voice cloning, music & sound effect generation.Extend ComfyUIPlugins, custom nodes, and extensions for ComfyUI & friends.Generate imagesHigh-end diffusion models — SDXL, Flux, SD3, Hunyuan.Generate videoLocal & cloud video diffusion: Wan, Hunyuan, CogVideoX, LTX.Run LLMs locallyRun large language models on your own hardware.Scale workloadsRun pipelines at scale — serverless GPUs, queue managers.Train modelsLoRA / DreamBooth / full fine-tune toolkits.Build node graphsNodal & graph-based AI workflow builders.
01 · This week
Trending AI tools
What students and creators are actually using this week.
Wan 2.2
Open-weight video diffusion from Alibaba.
Open Source 12–48 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- Workstation GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Ollama
One-command local LLM runtime.
Open SourceCPU-capable
Actually FreeNo SignupOpen SourceWatermark-Free
FLUX.1 [dev]
12B parameter open-weight diffusion model.
Open Source 8–24 GB VRAM
- Min VRAM
- 8 GB
- GPU class
- High-end GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Cline
Agentic coding in VS Code — reads, writes, runs, browses.
Open Source☁ Cloud · no GPU
Actually FreeNo SignupOpen SourceWatermark-Free
HunyuanVideo
13B open-weight cinematic text-to-video.
Open Source 24–48 GB VRAM
- Min VRAM
- 24 GB
- GPU class
- Workstation GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
02 · Free & open-weight
Best free / open-weight tools
OSS workflow engines, local runners, and open-weight models. You provide the GPU.
✓ Actually FreeOpen SourceNo SignupWatermark-Free
ComfyUI
The nodal workflow engine for serious diffusion.
Open Source 6–16 GB VRAM
- Min VRAM
- 6 GB
- GPU class
- Mid GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
llama.cpp
The C++ inference engine powering most local LLMs.
Open SourceCPU-capable
Actually FreeNo SignupOpen SourceWatermark-Free
ComfyUI-Manager
Install, update, and govern ComfyUI custom nodes.
Open Sourcevia comfyui
Actually FreeNo SignupOpen SourceWatermark-Free
FLUX.1 [dev]
12B parameter open-weight diffusion model.
Open Source 8–24 GB VRAM
- Min VRAM
- 8 GB
- GPU class
- High-end GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Ollama
One-command local LLM runtime.
Open SourceCPU-capable
Actually FreeNo SignupOpen SourceWatermark-Free
Transformers
The library every LLM ships against first.
Open Source 2–24 GB VRAM
- Min VRAM
- 2 GB
- GPU class
- Entry GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
03 · Fresh drops
New AI releases
Freshly launched or recently updated tools.
Wan 2.2
Open-weight video diffusion from Alibaba.
Open Source 12–48 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- Workstation GPU
- Quant
- FP16
Actually FreeNo SignupOpen SourceWatermark-Free
Kling 2.1
Cloud video gen with strong motion control.
Paid · from $10/mo☁ Cloud · no GPU
Watermark-FreeAPI
Magi-1
Autoregressive video diffusion at 24 GB.
Open Source 24–48 GB VRAM
- Min VRAM
- 24 GB
- GPU class
- Workstation GPU
- Quant
- FP8
Actually FreeNo SignupOpen SourceWatermark-Free
Runway Gen-4
Cinematic-quality cloud video generation.
Paid · from $15/mo☁ Cloud · no GPU
Watermark-FreeAPI
Hunyuan3D-2
Tencent's open 3D generator — multi-view, PBR, ready-to-use meshes.
Open Source 12–16 GB VRAM
- Min VRAM
- 12 GB
- GPU class
- High-end GPU
- Quant
- FP16
Actually FreeOpen SourceWatermark-FreeAPI
NVIDIA Cosmos
World-foundation models for physical AI.
Open Source 24–80 GB VRAM
- Min VRAM
- 24 GB
- GPU class
- Datacenter GPU
- Quant
- FP16
Actually FreeWatermark-FreeAPI
04 · Head-to-head
Side-by-side comparisons
Skip the directory rabbit hole — see the top picks compared directly.
Updated 2026-05-19
ComfyUI vs AUTOMATIC1111 vs InvokeAI
Which diffusion workflow engine should you commit to?
ComfyUI wins on flexibility and new-model support. A1111 wins on extension breadth. InvokeAI wins on production polish and canvas-first work.
Updated 2026-05-18
Wan 2.2 vs HunyuanVideo vs LTX-Video
Local video diffusion at three VRAM tiers.
Wan 2.2 is the new open-weight quality leader. Hunyuan still has the most cinematic motion. LTX-Video is the only one that's practical on a 16 GB card.
Updated 2026-05-15
Ollama vs LM Studio vs llama.cpp
The three serious ways to run LLMs locally.
Ollama is the fastest path to a working OpenAI-compatible server. LM Studio is the best GUI, especially on Apple Silicon. llama.cpp gives you full control and weirdest quants.
Updated 2026-05-13
Local video diffusion vs Runway Gen-4
When does paying per credit beat owning a 4090?
For occasional polished shots, Runway is cheaper than buying a workstation GPU. For iteration-heavy work or non-public projects, local Wan/Hunyuan pays for itself in months.
The thesis
Decisions, not directories.
We focus on the few tools worth your time — and tell you which ones are actually free. Bookmark us; we update weekly.