FR

FrontTerrain-HARP-1B

FrontterrainGGUF by frontterrainUpdated 2026-10-27
Download on Hugging Face
Parameters123M
Context512
Layers12
Architecturellama
Smallest quant0.2 GB
Quick answer

FrontTerrain-HARP-1B needs about 0.8 GB of memory at F16 (a 0.2 GB file) with an 8K context window, so it fits on a 4 GB GPU.

FrontTerrain HARP-1B, a 123‑M parameter llama model for systems programming

HARP-1B is a 123‑million parameter llama model trained from scratch by FrontTerrain. It is optimized for systems programming, mathematics, Python, and native FrontLang source code generation. The model is released under the apache‑2.0 license.

Good at

  • Systems programming code generation
  • Mathematics and Python code
  • FrontLang (.ft) source support

Limits

  • No vision support
  • Context limited to 512 tokens
  • Requires 0.8 GB VRAM at 8k context

Best for

  • Developers writing systems code
  • Educators teaching programming concepts
  • Tooling for FrontLang development

Who it’s for. Developers and researchers who need a lightweight language model focused on systems programming and FrontLang. The model is suitable for integration into IDEs and educational tools.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
F160.2 GB 0.8 GB
Fits Get file →

The only available quant is F16, 0.25 GB file size. It requires about 0.8 GB VRAM at 8k context. This is suitable for machines with at least 1 GB VRAM. No other quants are available.

What GPU do you need for FrontTerrain-HARP-1B?

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 F16 (0.2 GB file, about 0.8 GB in use).

Which FrontTerrain-HARP-1B quant should you choose?

F16 (0.2 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

Architecturellama
Parameters123M
Layers12
Trained context512 tokens
KV cache (fp16)9 KB per token

Source

Quantized byfrontterrain
Licenseapache-2.0
Downloads (HF)58,704
Last updated2026-10-27

Frequently asked questions about FrontTerrain-HARP-1B

How much VRAM does FrontTerrain-HARP-1B need?

About 0.8 GB at F16 with an 8K context window. That is the 0.2 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 FrontTerrain-HARP-1B on an 8 GB GPU?

Yes, with F16: the 0.2 GB file needs about 0.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of FrontTerrain-HARP-1B GGUF?

It depends on the quant: F16 is 0.2 GB.

What context length does FrontTerrain-HARP-1B support?

The GGUF header lists a trained context length of 512 tokens (512). The KV cache grows with the context you use: about 0.1 GB at 8K and 0.0 GB at 512.

Where can I download FrontTerrain-HARP-1B GGUF?

The files are on Hugging Face in the frontterrain/HARP-1B repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does FrontTerrain-HARP-1B 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 HARP-1B on 8 GB VRAM?

Yes, the F16 quant requires only 0.8 GB VRAM at 8k context, well within an 8 GB VRAM budget.

What context length does HARP-1B support?

The model was trained with a context of 512 tokens and is limited to that length.

Is HARP-1B suitable for vision tasks?

No, the model has no vision support and is focused solely on text and code generation.

Similar models

Browse all →

Other tiny models

All tiny models →