SA

Sarashina2.2 0.5b Instruct v0.1

SarashinaGGUF by mmngaUpdated 2025-03-05
Download on Hugging Face
Parameters793M
Context8K
Layers24
Architecturellama
Smallest quant0.3 GB
Quick answer

Sarashina2.2 0.5b Instruct v0.1 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 24 quants, from 0.3 GB (IQ1_S) to 0.8 GB (Q8_0).

Japanese instruction model, 0.793B parameters

Sarashina2.2 0.5b Instruct v0.1 is a 0.793B parameter autoregressive language model trained for Japanese instruction following. It supports up to 8192 context tokens and is released under the MIT license.

Good at

  • Japanese instruction following
  • Short context generation
  • Low‑memory inference

Limits

  • Limited safety training
  • No vision support
  • No other languages

Best for

  • Chatbots in Japanese
  • Japanese text summarization
  • Japanese code generation

Who it’s for. Developers building Japanese‑focused LLM applications who need a lightweight model with instruction capabilities. The model is suitable for local deployment on modest hardware.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
IQ1_S0.3 GB 1.3 GB
Fits Get file →
IQ1_M0.3 GB 1.3 GB
Fits Get file →
IQ2_XXS0.3 GB 1.3 GB
Fits Get file →
IQ2_XS0.3 GB 1.3 GB
Fits Get file →
IQ2_S0.3 GB 1.4 GB
Fits Get file →
IQ2_M0.4 GB 1.4 GB
Fits Get file →
Q2_K0.4 GB 1.4 GB
Fits Get file →
IQ3_XXS0.4 GB 1.4 GB
Fits Get file →
IQ3_XS0.4 GB 1.4 GB
Fits Get file →
Q3_K_S0.4 GB 1.4 GB
Fits Get file →
IQ3_S0.4 GB 1.4 GB
Fits Get file →
IQ3_M0.4 GB 1.4 GB
Fits Get file →
Q3_K_M0.4 GB 1.4 GB
Fits Get file →
Q3_K_L0.5 GB 1.5 GB
Fits Get file →
IQ4_XS0.5 GB 1.5 GB
Fits Get file →
Q4_00.5 GB 1.5 GB
Fits Get file →
IQ4_NL0.5 GB 1.5 GB
Fits Get file →
Q4_K_S0.5 GB 1.5 GB
Fits Get file →
Q4_K_M0.5 GB 1.5 GB
Fits Get file →
Q5_00.6 GB 1.6 GB
Fits Get file →
Q5_K_S0.6 GB 1.6 GB
Fits Get file →
Q5_K_M0.6 GB 1.6 GB
Fits Get file →
Q6_K0.7 GB 1.7 GB
Fits Get file →
Q8_00.8 GB 1.8 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 IQ3_S (0.41 GB file, ~1.4 GB memory). For 2 GB VRAM, use IQ2_XS (0.33 GB file, ~1.3 GB memory). Estimates are approximate.

What GPU do you need for Sarashina2.2 0.5b Instruct v0.1?

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 Q8_0 (0.8 GB file, about 1.8 GB in use).

Which Sarashina2.2 0.5b Instruct v0.1 quant should you choose?

Q4_K_M is the usual starting point for Sarashina2.2 0.5b Instruct v0.1: 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 IQ1_S at 0.3 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

Architecturellama
Parameters793M
Layers24
Trained context8,192 tokens
KV cache (fp16)60 KB per token

Source

Quantized bymmnga
FamilySarashina
Licensemit
Downloads (HF)1,429
Last updated2025-03-05

Frequently asked questions about Sarashina2.2 0.5b Instruct v0.1

How much VRAM does Sarashina2.2 0.5b Instruct v0.1 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 Instruct v0.1 on an 8 GB GPU?

Yes, with Q8_0: the 0.8 GB file needs about 1.8 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 Instruct v0.1 GGUF?

It depends on the quant: IQ1_S is 0.3 GB, IQ1_M is 0.3 GB, IQ2_XXS is 0.3 GB, IQ2_XS is 0.3 GB, IQ2_S is 0.3 GB, IQ2_M is 0.4 GB and 18 more.

What context length does Sarashina2.2 0.5b Instruct v0.1 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 Instruct v0.1 GGUF?

The files are on Hugging Face in the mmnga/sarashina2.2-0.5b-instruct-v0.1-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 Instruct v0.1 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, the Q4_K_M quant requires about 1.5 GB VRAM at 8 k context, which fits in 8 GB.

Does this model support English?

The model is trained for Japanese; English performance is limited and not guaranteed.

Is there a safety filter?

The model has limited safety training and may produce inaccurate or biased outputs; tuning is recommended.

Other GGUFs of sarashina2.2-0.5b-instruct-v0.1

Similar models

Browse all →

More from mmnga

All by mmnga →

More Sarashina models

All Sarashina →

Other tiny models

All tiny models →