laya needs about 1.7 GB of memory at Q4_K_M (a 0.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 4 quants, from 0.3 GB (Q4_K_M) to 0.8 GB (F16).
Multilingual, non‑autoregressive decision model with 8,192‑token context
Laya is a 0.395B modern‑bert model that returns typed answers with calibrated probabilities in a single forward pass. It supports over 100 languages and reads up to 8,192 tokens. The model is released under Apache‑2.0 and can be used with llama.cpp or the laya Python package.
Good at
- Fast single‑pass decision making
- Accurate probability calibration
- Multilingual support
Limits
- No text generation, only decision outputs
- Requires separate head for full inference
- Limited long‑document accuracy beyond 4,000 tokens
Best for
- Customer support ticket classification
- Email routing decisions
- Multilingual intent detection
Who it’s for. Developers building decision‑making systems that need fast, probabilistic answers across many languages. Suitable for integration in web services, chatbots, or internal tooling where hallucination is undesirable.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 0.3 GB | 1.7 GB | Fits | Get file → | |
| Q6_K | 0.3 GB | 1.8 GB | Fits | Get file → | |
| Q8_0 | 0.4 GB | 1.9 GB | Fits | Get file → | |
| F16 | 0.8 GB | 2.2 GB | Fits | Get file → |
For 8 GB VRAM, use Q4_K_M (0.27 GB file, ~1.7 GB VRAM at 8 k context). For 12 GB VRAM, Q6_K (0.34 GB, ~1.8 GB). For 16 GB or more, Q8_0 (0.42 GB, ~1.9 GB) or the unquantized F16 (0.79 GB, ~2.2 GB). Memory estimates are approximate.
What GPU do you need for laya?
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 (0.8 GB file, about 2.2 GB in use).
Which laya quant should you choose?
Q4_K_M is the usual starting point for laya: it balances file size and quality well. Here it is a 0.3 GB file.
The Q8_0 file is 0.4 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 laya
How much VRAM does laya need?
About 1.7 GB at Q4_K_M with an 8K context window. That is the 0.3 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 laya on an 8 GB GPU?
Yes, with F16: the 0.8 GB file needs about 2.2 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of laya GGUF?
It depends on the quant: Q4_K_M is 0.3 GB, Q6_K is 0.3 GB, Q8_0 is 0.4 GB, F16 is 0.8 GB.
What context length does laya support?
The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 0.9 GB at 8K and 0.9 GB at 8K.
Where can I download laya GGUF?
The files are on Hugging Face in the fr0stbit3/laya-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 laya 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 Laya on 8 GB VRAM?
Yes, the Q4_K_M quantized file is 0.27 GB and requires about 1.7 GB VRAM at 8 k context.
Does Laya support image inputs?
No, Laya is a text‑only decision model and does not handle vision inputs.
Is the model fine‑tunable?
Yes, fine‑tuning on domain decisions improves accuracy; a notebook is available for Kaggle T4 GPUs.