OPEN MODELS / Language models (LLM)
Run granite3-dense:8b on a cloud GPU
granite3-dense:8b is a Language models (LLM) model by Ollama. See how much GPU memory it needs, which GPUs it fits on today and what it costs to run it yourself, billed per second.
From the project: The IBM Granite 2B and 8B models are designed to support tool-based use cases and support for retrieval augmented generation (RAG), streamlining code generation, translation and bug fixing.
GPUs it fits on today
About 7 GB of GPU memory: the model download (4-bit by default) plus working memory.
| GPU | Memory | Per hour | 10 hours | |
|---|---|---|---|---|
| GTX 1070 | 8 GB | $0.14 | $1.40 | Rent |
| RTX 3060 | 12 GB | $0.14 | $1.40 | Rent |
| GTX 1070 Ti | 8 GB | $0.16 | $1.60 | Rent |
| GTX 1080 | 8 GB | $0.16 | $1.60 | Rent |
| Q RTX 4000 | 8 GB | $0.16 | $1.60 | Rent |
Live starting prices from our catalog. You pay per second while the machine runs.
How to run granite3-dense:8b
- 01
Rent a GPU
Pick one of the GPUs above. Billing is per second; stop the machine and you stop paying.
- 02
Open a terminal or Jupyter
Connect over SSH or open Jupyter in your browser. PyTorch and CUDA are ready to use.
- 03
Install Ollama and run the model
Run the commands below. Ollama downloads the model and serves an OpenAI-compatible API on port 11434.
curl -fsSL https://ollama.com/install.sh | sh
ollama run granite3-dense:8bWhy run it yourself?
Your own LLM server has no per-token bill and no rate limit, and your data stays private. Pick a GPU by memory and serve the model with vLLM.
Questions
- How much GPU memory does granite3-dense:8b need?
- By our estimate about 7 GB for inference (8B parameters). Higher resolution, longer context or bigger batches need more.
- How much does it cost to run granite3-dense:8b?
- From $0.14 per hour on GTX 1070, billed per second. Ten hours cost about $1.40.
- Do I need a subscription?
- No. You pay only for the time your machine runs, from your USD balance. Stop it whenever you are done.
Try it on your own GPU in minutes
Pick a GPU, open Jupyter or SSH and run any open model. You pay per second and can stop anytime; no subscription, no setup fee.
- Per-second billing in USD
- PyTorch and CUDA ready in minutes
- Rent from ChatGPT or Claude too