Model Marketplace
Open models served by Ritual's distributed compute layer
online
LLaMA 2 30B
Meta
General-purpose large language model with strong reasoning.
Speed
42 t/s
Cost
0.0009 R/1K
online
Mistral 7B
Mistral AI
Fast, compact, instruction-tuned model. Great for routing.
Speed
96 t/s
Cost
0.0003 R/1K
online
Stable Diffusion XL
Stability AI
High-fidelity 1024px image synthesis on distributed GPUs.
Speed
14 t/s
Cost
0.0042 R/1K
degraded
GPT-J 6B
EleutherAI
Open-source autoregressive model, lightweight and cheap.
Speed
78 t/s
Cost
0.00025 R/1K