Browse open models by task and hardware.
Set your hardware to see model fit beside each result. Open an entry for license details, performance data, and sources.
All types
LLM
Image Model
Video Model
Audio Model
Embedding Model
Reranker Model
Multimodal Model
Agent Framework
Workflow Tool
Robotics System
Research Paper
Dataset
Benchmark
API Provider
Local Runtime
Hardware Tool
Demo
Sponsored Offer
Unknown
4 results, 25 in the catalog
Just browsing
≤16 GB GPU
24 GB GPU
32 GB GPU
Mac
Recommended
Newest
Top stamped
Lowest hardware
Name A to Z
IndexTTS-2.5
Audio Model
I load one reference voice, nudge the emotion and tempo faders, then cue the pronunciation. Done.
≥8 GB VRAMCustom
Grade A7 sources
MiDashengLM-Gen
Audio Model
I cue a voice, let the music enter, then hear the effects and room tone build around them. MiDashengLM-Gen keeps playing until its learned stop head lowers the needle.
Qwen3-1.7B backbone plus 16-layer audio DiT≥16 GB VRAMApache-2.0
Grade A5 sources
VocalRender
Audio Model
Like a conductor by the pond, hand it a written leaf, and a voice, produces a nice chirp!
Approximately 2.3B ARDM parameters (1.7B AR transformer + 0.6B DiT; AudioVAE separate)≥16 GB VRAMApache-2.0, allowed
Grade A7 sources
SymphonyGen
Audio Model
I sketch the harmony skeleton, then it be-boops and seats the orchestra bar by bar, making a catchy tune.
124M orchestration model + 87M harmony model≥8 GB VRAMMIT, allowed
Grade B4 sources