SPARK Summarization needs about 27.9 GB of memory at BF16 (a 24.5 GB file) with an 8K context window, so it fits on a 32 GB GPU.
SPARK Summarization is a 12.2B-parameter Mistral model in GGUF format, quantized by kistepAI. Its trained context length is 128K tokens. The repository has 1 quant file, from 24.5 GB (BF16) to 24.5 GB (BF16).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| BF16 | 24.5 GB | 27.9 GB | Too big | Get file → |
What GPU do you need for SPARK Summarization?
These figures add the file size, an fp16 KV cache at an 8K context window 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: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: the best quant that fits is BF16 (24.5 GB file, about 27.9 GB in use).
Which SPARK Summarization quant should you choose?
BF16 (24.5 GB) is the middle option in this repository. The unquantized model. It gives original quality and needs the most memory, so it is mainly for conversion or reference.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about SPARK Summarization
How much VRAM does SPARK Summarization need?
About 27.9 GB at BF16 with an 8K context window. That is the 24.5 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 SPARK Summarization on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 27.9 GB, so you would need a larger GPU or CPU offloading.
What is the file size of SPARK Summarization GGUF?
It depends on the quant: BF16 is 24.5 GB.
What context length does SPARK Summarization support?
The GGUF header lists a trained context length of 131,072 tokens (128K).
Where can I download SPARK Summarization GGUF?
The files are on Hugging Face in the kistepAI/SPARK-Summarization-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.