Sarashina2.2 0.5b needs about 1.5 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 12 quants, from 0.4 GB (Q2_K) to 1.6 GB (F16).
Sarashina2.2 0.5b – a 0.793B parameter Llama model
Sarashina2.2 0.5b is a 0.793B parameter Llama architecture model trained on 10 trillion tokens. It supports a 8192 token context window and is released under the MIT license. The model is not tuned for instruction following.
Good at
- Japanese language tasks
- Math and coding tasks
- General application tasks
Limits
- No instruction tuning
- No vision support
- Limited to 8192 token context
Best for
- Japanese QA
- Math problem solving
- Code generation
Who it’s for. Developers and researchers needing a small Japanese‑focused Llama model. The model is suitable for local inference with moderate GPU memory.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 0.4 GB | 1.4 GB | Fits | Get file → | |
| Q3_K_S | 0.4 GB | 1.4 GB | Fits | Get file → | |
| Q3_K_M | 0.4 GB | 1.4 GB | Fits | Get file → | |
| Q3_K_L | 0.5 GB | 1.5 GB | Fits | Get file → | |
| IQ4_XS | 0.5 GB | 1.5 GB | Fits | Get file → | |
| Q4_K_S | 0.5 GB | 1.5 GB | Fits | Get file → | |
| Q4_K_M | 0.5 GB | 1.5 GB | Fits | Get file → | |
| Q5_K_S | 0.6 GB | 1.6 GB | Fits | Get file → | |
| Q5_K_M | 0.6 GB | 1.6 GB | Fits | Get file → | |
| Q6_K | 0.7 GB | 1.7 GB | Fits | Get file → | |
| Q8_0 | 0.8 GB | 1.8 GB | Fits | Get file → | |
| F16 | 1.6 GB | 2.6 GB | Fits | Get file → |
For 8 GB VRAM, use Q4_K_M (0.53 GB file, ~1.5 GB memory). For 4 GB VRAM, use Q3_K_S (0.41 GB file, ~1.4 GB memory). For 2 GB VRAM, use Q2_K (0.37 GB file, ~1.4 GB memory). Estimates are approximate.
What GPU do you need for Sarashina2.2 0.5b?
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 (1.6 GB file, about 2.6 GB in use).
Which Sarashina2.2 0.5b quant should you choose?
Q4_K_M is the usual starting point for Sarashina2.2 0.5b: it balances file size and quality well. Here it is a 0.5 GB file.
The Q8_0 file is 0.8 GB, about 1.6 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.
If memory is tight, the smallest file is Q2_K at 0.4 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Sarashina2.2 0.5b
How much VRAM does Sarashina2.2 0.5b need?
About 1.5 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 Sarashina2.2 0.5b on an 8 GB GPU?
Yes, with F16: the 1.6 GB file needs about 2.6 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Sarashina2.2 0.5b GGUF?
It depends on the quant: Q2_K is 0.4 GB, Q3_K_S is 0.4 GB, Q3_K_M is 0.4 GB, Q3_K_L is 0.5 GB, IQ4_XS is 0.5 GB, Q4_K_S is 0.5 GB and 6 more.
What context length does Sarashina2.2 0.5b support?
The GGUF header lists a trained context length of 8,192 tokens (8K). The KV cache grows with the context you use: about 0.5 GB at 8K and 0.5 GB at 8K.
Where can I download Sarashina2.2 0.5b GGUF?
The files are on Hugging Face in the mradermacher/sarashina2.2-0.5b-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 Sarashina2.2 0.5b use?
Hugging Face lists the license as mit. 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 this model on 8 GB VRAM?
Yes, using Q4_K_M or Q4_K_S quantizations will fit within 8 GB VRAM with an estimated 1.5 GB memory usage.
Does this model support vision tasks?
No, the model does not include vision capabilities.
Is the model instruction‑tuned?
No, the base model is not instruction‑tuned and may generate inaccurate or biased outputs.