Kanana 1.5 15.7b A3B Instruct needs about 12.0 GB of memory at Q4_K_M (a 10.4 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 11 quants, from 6.5 GB (Q2_K) to 16.7 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
Kanana 1.5 15.7b A3B Instruct – a Mixture‑of‑Experts Llama model
Kanana 1.5 15.7b A3B Instruct is a 15.7 billion‑parameter Llama‑style model with a sparse MoE architecture. It supports 32 768‑token context and is optimized for instruction following. The model is available in multiple GGUF quantizations.
Good at
- Instruction following
- Tool usage
- Sparse inference efficiency
Limits
- No vision support
- License type other
- Requires 8 k context for memory estimates
Best for
- Chatbots with limited GPU memory
- Tool‑integrated assistants
- Cost‑effective inference
Who it’s for. Developers and researchers needing a large‑scale instruction model that can run on modest hardware. The MoE design allows efficient inference while maintaining strong performance.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 6.5 GB | 8.1 GB | Too big | Get file → | |
| Q3_K_S | 7.5 GB | 9.1 GB | Too big | Get file → | |
| Q3_K_M | 8.2 GB | 9.8 GB | Too big | Get file → | |
| Q3_K_L | 8.5 GB | 10.1 GB | Too big | Get file → | |
| IQ4_XS | 8.7 GB | 10.3 GB | Too big | Get file → | |
| Q4_K_S | 9.6 GB | 11.1 GB | Too big | Get file → | |
| Q4_K_M | 10.4 GB | 12.0 GB | Too big | Get file → | |
| Q5_K_S | 11.2 GB | 12.7 GB | Too big | Get file → | |
| Q5_K_M | 11.9 GB | 13.5 GB | Too big | Get file → | |
| Q6_K | 14.1 GB | 15.6 GB | Too big | Get file → | |
| Q8_0 | 16.7 GB | 18.3 GB | Too big | Get file → |
For 8 GB VRAM, use Q2_K (6.48 GB) or Q3_K_S (7.51 GB). For 12 GB VRAM, Q4_K_M (10.44 GB) is suitable. For 18 GB VRAM, Q8_0 (16.7 GB) provides the highest quality.
What GPU do you need for Kanana 1.5 15.7b A3B 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: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: the best quant that fits is Q4_K_M (10.4 GB file, about 12.0 GB in use).
- 16 GB VRAM: the best quant that fits is Q6_K (14.1 GB file, about 15.6 GB in use).
- 24 GB VRAM: the best quant that fits is Q8_0 (16.7 GB file, about 18.3 GB in use).
Which Kanana 1.5 15.7b A3B Instruct quant should you choose?
Q4_K_M is the usual starting point for Kanana 1.5 15.7b A3B Instruct: it balances file size and quality well. Here it is a 10.4 GB file.
The Q8_0 file is 16.7 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 6.5 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 1.5 15.7b A3B Instruct
How much VRAM does Kanana 1.5 15.7b A3B Instruct need?
About 12.0 GB at Q4_K_M with an 8K context window. That is the 10.4 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 1.5 15.7b A3B Instruct on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 8.1 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Kanana 1.5 15.7b A3B Instruct GGUF?
It depends on the quant: Q2_K is 6.5 GB, Q3_K_S is 7.5 GB, Q3_K_M is 8.2 GB, Q3_K_L is 8.5 GB, IQ4_XS is 8.7 GB, Q4_K_S is 9.6 GB and 5 more.
What context length does Kanana 1.5 15.7b A3B 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 1.5 15.7b A3B Instruct GGUF?
The files are on Hugging Face in the mradermacher/kanana-1.5-15.7b-a3b-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 1.5 15.7b A3B 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 1.5 15.7b A3B on 8 GB VRAM?
Yes, the Q2_K quant is 6.48 GB and the Q3_K_S quant is 7.51 GB, both fit within 8 GB VRAM.
Does this model support vision tasks?
No, the model does not include vision capabilities.
What context length does the model support?
The model supports a context length of 32 768 tokens.