SP

Sparkx25

SparkGGUF by pfeifferjUpdated 2026-09-13
Download on Hugging Face
Parameters4.11B
Context1024K
Layers36
Architecturespark2_5
Smallest quant1.5 GB
Quick answer

Sparkx25 needs about 3.2 GB of memory at its main file (a 1.5 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 1.5 GB to 1.8 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 4.8 GB.

Sparkx25 – a 4.1B‑parameter GGUF model with 1.5–1.8 GB file size

Sparkx25 is a 4.11‑billion‑parameter language model released under the Apache‑2.0 license. It offers a native 1M‑token context window and supports reasoning and tool use. The model is available in two GGUF quantizations, 1.54 GB and 1.8 GB, with estimated VRAM usage of 3.2 GB and 3.5 GB at 8 k context.

Good at

  • Reasoning
  • Tool use
  • Large context handling

Limits

  • No vision support
  • Requires at least 3.2 GB VRAM
  • No official benchmark scores provided

Best for

  • Local inference on laptops
  • Agent workflows
  • Coding assistance

Who it’s for. Developers and researchers who need a compact, high‑capacity model for local deployment. The model is suitable for users with limited GPU memory who still require a large context window.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—1.5 GB 3.2 GB
Fits Get file →
—1.8 GB 3.5 GB
Fits Get file →

For 8 GB VRAM, the 1.54 GB GGUF file is recommended, using about 3.2 GB of VRAM at 8 k context. If you have 12 GB VRAM, the 1.8 GB file can be used, consuming roughly 3.5 GB of VRAM. Both files are approximate and may vary with the inference build.

What GPU do you need for Sparkx25?

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 (1.8 GB file, about 3.5 GB in use).

Specifications

Architecture from the GGUF header

Architecturespark2_5
Parameters4.11B
Layers36
Trained context1,048,576 tokens
KV cache (fp16)144 KB per token

Source

Quantized bypfeifferj
FamilySpark
Licenseapache-2.0
Downloads (HF)845
Last updated2026-09-13

Frequently asked questions about Sparkx25

How much VRAM does Sparkx25 need?

About 3.2 GB at its main file with an 8K context window. That is the 1.5 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 Sparkx25 on an 8 GB GPU?

Yes, with the smaller files: the 1.8 GB file needs about 3.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Sparkx25 GGUF?

It depends on the quant: file is 1.5 GB, file is 1.8 GB.

What context length does Sparkx25 support?

The GGUF header lists a trained context length of 1,048,576 tokens (1024K). The KV cache grows with the context you use: about 1.2 GB at 8K and 4.8 GB at 32K.

Where can I download Sparkx25 GGUF?

The files are on Hugging Face in the pfeifferj/Spark-X2.5-4B-GSQ-RCO-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Sparkx25 use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run Sparkx25 on 8 GB VRAM?

Yes, the 1.54 GB GGUF file uses about 3.2 GB of VRAM at 8 k context, which fits within an 8 GB GPU.

Does Sparkx25 support vision tasks?

No, the model does not include vision capabilities.

What license applies to Sparkx25?

The model is released under the Apache‑2.0 license.

Other GGUFs of Spark-X2.5-4B

Similar models

Browse all →

More from pfeifferj

All by pfeifferj →

More Spark models

All Spark →

Other small models

All small models →