Models

Popular models from Hugging Face that GridPup can serve, grouped by family and size, with the GPU memory each one needs and the cheapest machine that fits it right now. Open a family to compare its sizes and versions (the original, quantized and ONNX versions need less GPU memory) and the GPUs for each.

1,054 families · catalog updated 12:33 UTC

• pygmalion by notstoicText generation · 1 model · ↓ 20613B
  Smallest needs12.4 GB GPU
  From$0.14/hRTX A4000
• Lumimaid-v0.2 by NeverSleepText generation · 1 model · ↓ 20412B
  Smallest needs30.9 GB GPU
  From$0.28/h4x RTX 3060
• Carnice by kai-osText generation · 1 model · ↓ 1949B
  Smallest needs23.3 GB GPU
  From$0.23/hRTX 3090
• MythoMax-L2 by GrypheText generation · 1 model · ↓ 19313B
  Smallest needs14 GB GPU
  From$0.14/hRTX A4000
• llama-3-firefunction by fireworks-aiText generation · 1 model · ↓ 178~70.6B
  Smallest needs162 GB GPU
  From$1.84/h4x RTX 4090

Model details come from Hugging Face and are refreshed daily. GPU memory is GridPup's estimate: the weights, room for an 8,192-token conversation (chat models) and the runtime's own memory. Large models are split across the GPUs of one machine ("4x RTX 4090"). Only machines with a Vast reliability above 95% that download at 400 Mbit/s or more are counted.