Hunyuan-A13B-Instruct needs about 50.4 GB of memory at Q4_K_M (a 48.8 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 26 quants, from 22.5 GB (TQ1_0) to 160.8 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 4.3 GB.
Hunyuan-A13B-Instruct: 80B total, 13B active, 256K context, MoE architecture
Hunyuan-A13B-Instruct is a large language model built on a fine‑grained Mixture‑of‑Experts architecture. It contains 80 billion parameters, of which 13 billion are active during inference. The model supports a 256 K token context window and is optimized for efficient inference with multiple quantization options.
Good at
- Large‑context reasoning
- Efficient inference with MoE
- Hybrid fast/slow thinking
Limits
- No vision support
- Requires 256 K context for full capability
- License is other
Best for
- Long‑form content generation
- Research prototypes with limited GPU memory
- Agent‑style task orchestration
Who it’s for. Researchers and developers needing high‑performance language modeling in resource‑constrained settings. The model is suitable for academic experiments and cost‑effective deployment scenarios.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| TQ1_0 | 22.5 GB | 24.1 GB | Too big | Get file → | |
| IQ1_S | 24.6 GB | 26.2 GB | Too big | Get file → | |
| IQ1_M | 25.6 GB | 27.1 GB | Too big | Get file → | |
| IQ2_XXS | 27.6 GB | 29.2 GB | Too big | Get file → | |
| IQ2_M | 28.8 GB | 30.4 GB | Too big | Get file → | |
| Q2_K | 29.4 GB | 31.0 GB | Too big | Get file → | |
| Q2_K_L | 29.4 GB | 31.0 GB | Too big | Get file → | |
| Q2_K_XL | 31.0 GB | 32.6 GB | Too big | Get file → | |
| IQ3_XXS | 34.2 GB | 35.8 GB | Too big | Get file → | |
| Q3_K_S | 34.8 GB | 36.4 GB | Too big | Get file → | |
| Q3_K_XL | 36.3 GB | 37.9 GB | Too big | Get file → | |
| Q3_K_M | 38.6 GB | 40.1 GB | Too big | Get file → | |
| IQ4_XS | 42.9 GB | 44.5 GB | Too big | Get file → | |
| IQ4_NL | 45.4 GB | 47.0 GB | Too big | Get file → | |
| Q4_0 | 45.6 GB | 47.2 GB | Too big | Get file → | |
| Q4_K_S | 45.8 GB | 47.4 GB | Too big | Get file → | |
| Q4_K_XL | 46.7 GB | 48.3 GB | Too big | Get file → | |
| Q4_K_M | 48.8 GB | 50.4 GB | Too big | Get file → | |
| Q4_1 | 50.4 GB (2 parts) | 52.0 GB | Too big | Get file → | |
| Q5_K_S | 55.4 GB (2 parts) | 56.9 GB | Too big | Get file → | |
| Q5_K_M | 57.1 GB (2 parts) | 58.7 GB | Too big | Get file → | |
| Q5_K_XL | 57.6 GB (2 parts) | 59.2 GB | Too big | Get file → | |
| Q6_K_XL | 70.5 GB (2 parts) | 72.1 GB | Too big | Get file → | |
| Q8_0 | 85.4 GB (2 parts) | 87.0 GB | Too big | Get file → | |
| Q8_K_XL | 97.9 GB (2 parts) | 99.4 GB | Too big | Get file → | |
| BF16 | 160.8 GB (4 parts) | 162.4 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_0 (file size 45.6 GB, estimated 47.2 GB memory at 8 k context). For 16 GB VRAM, Q4_K_M (48.78 GB, 50.4 GB memory) or Q5_K_S (55.37 GB, 56.9 GB memory). For 24 GB VRAM, Q5_K_M (57.12 GB, 58.7 GB memory).
What GPU do you need for Hunyuan-A13B-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: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: the best quant that fits is Q2_K (29.4 GB file, about 31.0 GB in use).
- 48 GB VRAM: the best quant that fits is Q4_K_S (45.8 GB file, about 47.4 GB in use).
Which Hunyuan-A13B-Instruct quant should you choose?
Q4_K_M is the usual starting point for Hunyuan-A13B-Instruct: it balances file size and quality well. Here it is a 48.8 GB file.
The Q8_0 file is 85.4 GB, about 1.8 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 TQ1_0 at 22.5 GB, with the quality trade-offs that come with fewer bits. Lower bit counts mean smaller files and more quality loss.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Hunyuan-A13B-Instruct
How much VRAM does Hunyuan-A13B-Instruct need?
About 50.4 GB at Q4_K_M with an 8K context window. That is the 48.8 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 Hunyuan-A13B-Instruct on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 24.1 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Hunyuan-A13B-Instruct GGUF?
It depends on the quant: TQ1_0 is 22.5 GB, IQ1_S is 24.6 GB, IQ1_M is 25.6 GB, IQ2_XXS is 27.6 GB, IQ2_M is 28.8 GB, Q2_K is 29.4 GB and 20 more.
What context length does Hunyuan-A13B-Instruct support?
The GGUF header lists a trained context length of 262,144 tokens (256K). 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 Hunyuan-A13B-Instruct GGUF?
The files are on Hugging Face in the unsloth/Hunyuan-A13B-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 Hunyuan-A13B-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 Hunyuan-A13B-Instruct on 8 GB VRAM?
Use the Q4_0 quantization; it requires approximately 47.2 GB memory at 8 k context, which is beyond 8 GB VRAM, so it is not feasible without further optimization.
Does the model support image inputs?
No, the model has no vision capabilities.
What is the maximum context length?
The model supports a 256 K token context window.