How much GPU memory does Bio-Medical-Llama-3-8B need?

About 21.5 GB for conversations of up to 8,192 tokens: 16.06 GB of weights, the cache for one full-length conversation and the runtime's own memory (an estimate: its settings are behind its licence).

What is the cheapest GPU to run Bio-Medical-Llama-3-8B?

Right now one RTX 3090 at $0.19 an hour; RTX 5090 at $0.71 an hour is the best value for its speed. Prices are live offers from GPU clouds, checked every 15 minutes.

Can Bio-Medical-Llama-3-8B run on an RTX 4090?

Yes: it needs about 21.5 GB and an RTX 4090 has 24 GB.

Models / Bio-Medical-Llama-3

ContactDoctor/Bio-Medical-Llama-3-8B

Text generation · served with vLLM · ↓ 84 downloads a month · View on Hugging Face

ContactDoctor/Bio-Medical-Llama-3-8B needs about 21.5 GB of GPU memory for conversations of up to 8,192 tokens (an estimate: its settings are behind its licence). It runs on one RTX 3090 from $0.19/h right now; best value: RTX 5090 at $0.71/h.

The model's publisher asks users to accept its licence (other). GridPup cannot download this one yet: choose an ungated version below.

Weights: 16 GB
GPU memory needed: about 21.5 GB (estimate)
Conversation length: 8,192 tokens
Memory · disk: 16 GB · 58 GB
Architecture: LlamaForCausalLM
Licence: other

Hardware and deploy

The publisher has not let GridPup download this licence-gated model yet. An ungated version of it (above) can be deployed.

Questions

How much GPU memory does Bio-Medical-Llama-3-8B need?: About 21.5 GB for conversations of up to 8,192 tokens: 16.06 GB of weights, the cache for one full-length conversation and the runtime's own memory (an estimate: its settings are behind its licence).
What is the cheapest GPU to run Bio-Medical-Llama-3-8B?: Right now one RTX 3090 at $0.19 an hour; RTX 5090 at $0.71 an hour is the best value for its speed. Prices are live offers from GPU clouds, checked every 15 minutes.
Can Bio-Medical-Llama-3-8B run on an RTX 4090?: Yes: it needs about 21.5 GB and an RTX 4090 has 24 GB.

Loading…