jevling 0.8b v1 needs about 1.4 GB of memory at Q4_K_M (a 0.5 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 4 quants, from 0.5 GB (Q4_K_M) to 1.5 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.6 GB.
Jevling‑0.8B‑v1: a lightweight System One decision model
Jevling‑0.8B‑v1 is a 0.752 billion‑parameter model fine‑tuned from Qwen/Qwen3.5‑0.8B for on‑device use. It answers typed questions in a single forward pass with no text generation. The model is optimized for 8 GB RAM and supports Traditional‑Chinese kiosk transcripts.
Good at
- Fast single‑pass decision answering
- No text generation needed
- Calibrated probability outputs
Limits
- No vision support
- Requires maintained llama.cpp branch
- Limited to System One format
Best for
- Kiosk ordering systems
- On‑device decision support
- Rapid multi‑question inference
Who it’s for. Developers building lightweight on‑device decision systems that need calibrated probabilities without generating text. The model is suitable for environments with limited memory and where Traditional‑Chinese support is required.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 0.5 GB | 1.4 GB | Fits | Get file → | |
| Q8_0 | 0.8 GB | 1.7 GB | Fits | Get file → | |
| Q8_0 | 1.1 GB | 2.0 GB | Fits | Get file → | |
| BF16 | 1.5 GB | 2.4 GB | Fits | Get file → |
For 8 GB VRAM, use Q4_K_M (0.53 GB file, ~1.4 GB memory at 8 k context). If more VRAM is available, Q8_0 (0.81 GB file, ~1.7 GB memory) offers higher precision. BF16 (1.52 GB file, ~2.4 GB memory) is suitable for larger memory budgets.
What GPU do you need for jevling 0.8b v1?
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 BF16 (1.5 GB file, about 2.4 GB in use).
Which jevling 0.8b v1 quant should you choose?
Q4_K_M is the usual starting point for jevling 0.8b v1: it balances file size and quality well. Here it is a 0.5 GB file.
The Q8_0 file is 0.8 GB, about 1.5 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about jevling 0.8b v1
How much VRAM does jevling 0.8b v1 need?
About 1.4 GB at Q4_K_M with an 8K context window. That is the 0.5 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 jevling 0.8b v1 on an 8 GB GPU?
Yes, with BF16: the 1.5 GB file needs about 2.4 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of jevling 0.8b v1 GGUF?
It depends on the quant: Q4_K_M is 0.5 GB, Q8_0 is 0.8 GB, Q8_0 is 1.1 GB, BF16 is 1.5 GB.
What context length does jevling 0.8b v1 support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.6 GB at 32K.
Where can I download jevling 0.8b v1 GGUF?
The files are on Hugging Face in the BricksDisplay/jevling-0.8b-v1-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 jevling 0.8b v1 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 Jevling‑0.8B‑v1 on 8 GB VRAM?
Yes, the Q4_K_M quantization uses 0.53 GB file and requires about 1.4 GB memory at 8 k context, fitting within 8 GB VRAM.
Does the model support vision inputs?
No, Jevling‑0.8B‑v1 does not support vision; it only processes text in the System One format.
Which llama.cpp branch is required?
Use the maintained system‑one branch (feat/system‑one) in mybigday/system‑one‑llama.cpp; stock llama.cpp cannot handle typed questions.