baseline scalable llama needs about 2.1 GB of memory at its main file (a 1.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. Longer context needs more: at 32K tokens the KV cache alone adds about 1.1 GB.
baseline scalable llama is a 1.24B-parameter Llama model in GGUF format, quantized by hayitsmaddy. Its trained context length is 128K tokens. The repository has 1 quant file, from 1.3 GB to 1.3 GB.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 1.3 GB | 2.1 GB | Fits | Get file → |
What GPU do you need for baseline scalable llama?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: the best quant that fits is the main file (1.3 GB file, about 2.1 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about baseline scalable llama
How much VRAM does baseline scalable llama need?
About 2.1 GB at its main file with an 8K context window. That is the 1.3 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run baseline scalable llama on an 8 GB GPU?
Yes, with the smaller files: the 1.3 GB file needs about 2.1 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of baseline scalable llama GGUF?
It depends on the quant: file is 1.3 GB.
What context length does baseline scalable llama support?
The GGUF header lists a trained context length of 131,072 tokens (128K). The KV cache grows with the context you use: about 0.3 GB at 8K and 1.1 GB at 32K.
Where can I download baseline scalable llama GGUF?
The files are on Hugging Face in the hayitsmaddy/baseline_scalable_llama repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.