Kanana 2 3b Instruct needs about 3.7 GB of memory at Q4_K_M (a 2.2 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 12 quants, from 1.4 GB (Q2_K) to 7.0 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
Kanana 2 3b Instruct – a 3.51B parameter instruction‑tuned model
Kanana 2 3b Instruct is a 3.51B parameter model based on the Qwen3 architecture. It is trained for instruction following and reasoning, with a native context length of 32,768 tokens. The model is available in multiple GGUF quantizations.
Good at
- Instruction following
- Reasoning
- Long‑context support
Limits
- No vision support
- License is other
- No explicit language list
Best for
- Chatbots needing long context
- Instruction‑driven applications
- On‑device inference with limited VRAM
Who it’s for. Developers and researchers who need a compact yet capable instruction‑tuned model for Korean and other languages. The model is suitable for on‑device deployment where memory is constrained.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 1.4 GB | 3.0 GB | Fits | Get file → | |
| Q3_K_S | 1.6 GB | 3.2 GB | Fits | Get file → | |
| Q3_K_M | 1.8 GB | 3.4 GB | Fits | Get file → | |
| Q3_K_L | 1.9 GB | 3.5 GB | Fits | Get file → | |
| IQ4_XS | 2.0 GB | 3.6 GB | Fits | Get file → | |
| Q4_K_S | 2.1 GB | 3.7 GB | Fits | Get file → | |
| Q4_K_M | 2.2 GB | 3.7 GB | Fits | Get file → | |
| Q5_K_S | 2.5 GB | 4.0 GB | Fits | Get file → | |
| Q5_K_M | 2.5 GB | 4.1 GB | Fits | Get file → | |
| Q6_K | 2.9 GB | 4.5 GB | Fits | Get file → | |
| Q8_0 | 3.7 GB | 5.3 GB | Fits | Get file → | |
| F16 | 7.0 GB | 8.6 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_K_M or Q4_K_S (est. 3.7 GB at 8 k context). For 4 GB VRAM, Q3_K_M or Q3_K_S (est. 3.4 GB). For 2 GB VRAM, Q2_K (est. 3 GB). Larger VRAM allows higher‑quality quants such as Q5_K_M or F16.
What GPU do you need for Kanana 2 3b Instruct?
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 Q5_K_S (2.5 GB file, about 4.0 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (3.7 GB file, about 5.3 GB in use).
- 8 GB VRAM: the best quant that fits is Q8_0 (3.7 GB file, about 5.3 GB in use).
- 12 GB VRAM: the best quant that fits is F16 (7.0 GB file, about 8.6 GB in use).
Which Kanana 2 3b Instruct quant should you choose?
Q4_K_M is the usual starting point for Kanana 2 3b Instruct: it balances file size and quality well. Here it is a 2.2 GB file.
The Q8_0 file is 3.7 GB, about 1.7 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 1.4 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 Kanana 2 3b Instruct
How much VRAM does Kanana 2 3b Instruct need?
About 3.7 GB at Q4_K_M with an 8K context window. That is the 2.2 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 Kanana 2 3b Instruct on an 8 GB GPU?
Yes, with Q8_0: the 3.7 GB file needs about 5.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Kanana 2 3b Instruct GGUF?
It depends on the quant: Q2_K is 1.4 GB, Q3_K_S is 1.6 GB, Q3_K_M is 1.8 GB, Q3_K_L is 1.9 GB, IQ4_XS is 2.0 GB, Q4_K_S is 2.1 GB and 6 more.
What context length does Kanana 2 3b Instruct support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 1.1 GB at 8K and 4.3 GB at 32K.
Where can I download Kanana 2 3b Instruct GGUF?
The files are on Hugging Face in the mradermacher/kanana-2-3b-instruct-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 Kanana 2 3b Instruct use?
Hugging Face lists the license as other. 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 Kanana 2 3b Instruct on 8 GB VRAM?
Yes, using Q4_K_M or Q4_K_S quantizations which require approximately 3.7 GB of VRAM at 8 k context.
Does this model support vision tasks?
No, the model does not include vision capabilities.
What context length does it support?
It natively supports up to 32,768 tokens.