hon9kon9ize Cantonese Llama 2 7B preview20240625 needs about 8.9 GB of memory at Q4_K_M (a 4.1 GB file) with an 8K context window, so it fits on a 12 GB GPU. This repository has 22 quants, from 2.5 GB (Q2_K) to 7.2 GB (Q8_0).
Cantonese Llama‑2 7B preview – 6.74B parameters
Cantonese Llama‑2 7B preview is a 6.74B‑parameter model built on Llama‑2 7B architecture. It is trained to understand and generate Cantonese text, including slang, colloquials, and internet terms. The model is released under CC‑BY‑SA‑4.0.
Good at
- Cantonese text generation
- Cantonese slang and colloquials
- Internet‑style Cantonese
Limits
- Preview version may hallucinate
- No vision support
- Requires 8k context for estimates
Best for
- Chatbots in Cantonese
- Content creation in Cantonese
- Language learning tools
Who it’s for. Developers building Cantonese‑language applications who need a lightweight LLM. Researchers exploring Cantonese NLP and community contributors to Hong Kong AI projects.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 2.5 GB | 5.2 GB | Fits | Get file → | |
| IQ3_XS | 2.8 GB | 5.4 GB | Fits | Get file → | |
| IQ3_S | 2.9 GB | 5.6 GB | Fits | Get file → | |
| Q3_K_S | 2.9 GB | 5.6 GB | Fits | Get file → | |
| IQ3_M | 3.1 GB | 5.8 GB | Fits | Get file → | |
| Q3_K | 3.3 GB | 5.9 GB | Fits | Get file → | |
| Q3_K_M | 3.3 GB | 5.9 GB | Fits | Get file → | |
| Q3_K_L | 3.6 GB | 6.2 GB | Fits | Get file → | |
| IQ4_XS | 3.6 GB | 6.3 GB | Fits | Get file → | |
| Q4_0 | 3.8 GB | 6.5 GB | Fits | Get file → | |
| IQ4_NL | 3.8 GB | 6.5 GB | Fits | Get file → | |
| Q4_K_S | 3.9 GB | 6.5 GB | Fits | Get file → | |
| Q4_K | 4.1 GB | 6.7 GB | Fits | Get file → | |
| Q4_K_M | 4.1 GB | 6.7 GB | Fits | Get file → | |
| Q4_1 | 4.2 GB | 6.9 GB | Fits | Get file → | |
| Q5_0 | 4.7 GB | 7.3 GB | Tight | Get file → | |
| Q5_K_S | 4.7 GB | 7.3 GB | Tight | Get file → | |
| Q5_K | 4.8 GB | 7.4 GB | Tight | Get file → | |
| Q5_K_M | 4.8 GB | 7.4 GB | Tight | Get file → | |
| Q5_1 | 5.1 GB | 7.7 GB | Tight | Get file → | |
| Q6_K | 5.5 GB | 8.2 GB | Too big | Get file → | |
| Q8_0 | 7.2 GB | 9.8 GB | Too big | Get file → |
For 8 GB VRAM, use Q4_0 (file size 3.83 GB, estimated 8.6 GB memory). For 12 GB VRAM, Q8_0 (file size 7.16 GB, estimated 12 GB memory) provides the highest quality. Lower‑bit quants like Q2_K give the smallest file size but higher memory usage.
What GPU do you need for hon9kon9ize Cantonese Llama 2 7B preview20240625?
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: the best quant that fits is IQ3_M (3.1 GB file, about 7.9 GB in use).
- 12 GB VRAM: the best quant that fits is Q8_0 (7.2 GB file, about 12.0 GB in use).
Which hon9kon9ize Cantonese Llama 2 7B preview20240625 quant should you choose?
Q4_K_M is the usual starting point for hon9kon9ize Cantonese Llama 2 7B preview20240625: it balances file size and quality well. Here it is a 4.1 GB file.
The Q8_0 file is 7.2 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 Q2_K at 2.5 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 hon9kon9ize Cantonese Llama 2 7B preview20240625
How much VRAM does hon9kon9ize Cantonese Llama 2 7B preview20240625 need?
About 8.9 GB at Q4_K_M with an 8K context window. That is the 4.1 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 hon9kon9ize Cantonese Llama 2 7B preview20240625 on an 8 GB GPU?
Yes, with IQ3_M: the 3.1 GB file needs about 7.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of hon9kon9ize Cantonese Llama 2 7B preview20240625 GGUF?
It depends on the quant: Q2_K is 2.5 GB, IQ3_XS is 2.8 GB, IQ3_S is 2.9 GB, Q3_K_S is 2.9 GB, IQ3_M is 3.1 GB, Q3_K is 3.3 GB and 16 more.
What context length does hon9kon9ize Cantonese Llama 2 7B preview20240625 support?
The GGUF header lists a trained context length of 4,096 tokens (4K). The KV cache grows with the context you use: about 4.3 GB at 8K and 2.1 GB at 4K.
Where can I download hon9kon9ize Cantonese Llama 2 7B preview20240625 GGUF?
The files are on Hugging Face in the RichardErkhov/hon9kon9ize_-_Cantonese-Llama-2-7B-preview20240625-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 hon9kon9ize Cantonese Llama 2 7B preview20240625 use?
Hugging Face lists the license as cc-by-sa-4.0. 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_0 quant uses about 8.6 GB memory at 8k context, which fits within 8 GB VRAM with some headroom.
Does this model support vision tasks?
No, it is a text‑generation model only.
What license applies to this model?
The model is released under CC‑BY‑SA‑4.0, allowing reuse with attribution and share‑alike.