CA

CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged

CantonesellmGGUF by mradermacherUpdated 2026-10-12
Download on Hugging Face
Parameters30.5B
Context40K
Layers48
Architectureqwen3moe
Smallest quant11.3 GB
Quick answer

CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged needs about 19.9 GB of memory at Q4_K_M (a 18.6 GB file) with an 8K context window, so it fits on a 24 GB GPU. This repository has 11 quants, from 11.3 GB (Q2_K) to 32.5 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 3.2 GB.

CantoneseLLM v2.0 30B A3B – a 30.5B‑parameter mixture‑of‑experts model

CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged is a 30.5B‑parameter model built on Qwen3‑30B‑A3B‑Base. It is a mixture‑of‑experts with 3.3B active parameters and 40960 trained context tokens. The model focuses on Cantonese instruction following and reasoning.

Good at

  • Cantonese instruction following
  • Reasoning in Cantonese
  • Large context handling

Limits

  • No vision support
  • Requires 8k context for memory estimates
  • Apache‑2.0 license

Best for

  • Chatbots needing Cantonese knowledge
  • Research on low‑resource language reasoning
  • Fine‑tuning Cantonese instruction models

Who it’s for. Developers and researchers who need a Cantonese‑knowledgeable instruct model and are willing to accept that the chain‑of‑thought is in Written Chinese. It is suitable for reproducing paper ablations or for use cases where Cantonese output is required.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
Q2_K11.3 GB 12.6 GB
Too big Get file →
Q3_K_S13.3 GB 14.6 GB
Too big Get file →
Q3_K_M14.7 GB 16.0 GB
Too big Get file →
Q3_K_L15.9 GB 17.2 GB
Too big Get file →
IQ4_XS16.6 GB 17.9 GB
Too big Get file →
Q4_K_S17.5 GB 18.8 GB
Too big Get file →
Q4_K_M18.6 GB 19.9 GB
Too big Get file →
Q5_K_S21.1 GB 22.4 GB
Too big Get file →
Q5_K_M21.7 GB 23.0 GB
Too big Get file →
Q6_K25.1 GB 26.4 GB
Too big Get file →
Q8_032.5 GB 33.8 GB
Too big Get file →

For 8 GB VRAM, the Q2_K quant (file size 11.26 GB, estimated 12.6 GB memory at 8k context) is recommended. If more VRAM is available, Q4_K_M (18.56 GB, 19.9 GB memory) offers higher quality. For the highest quality within a 32 GB VRAM budget, Q8_0 (32.48 GB, 33.8 GB memory) can be used.

What GPU do you need for CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged?

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: the best quant that fits is Q3_K_M (14.7 GB file, about 16.0 GB in use).
  • 24 GB VRAM: the best quant that fits is Q5_K_M (21.7 GB file, about 23.0 GB in use).
  • 32 GB VRAM: the best quant that fits is Q6_K (25.1 GB file, about 26.4 GB in use).
  • 48 GB VRAM: the best quant that fits is Q8_0 (32.5 GB file, about 33.8 GB in use).

Which CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged quant should you choose?

Q4_K_M is the usual starting point for CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged: it balances file size and quality well. Here it is a 18.6 GB file.

The Q8_0 file is 32.5 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 11.3 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

Architectureqwen3moe
Parameters30.5B (approximate, mixture of experts)
Active parameters3B
Layers48
Trained context40,960 tokens
KV cache (fp16)96 KB per token

Frequently asked questions about CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged

How much VRAM does CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged need?

About 19.9 GB at Q4_K_M with an 8K context window. That is the 18.6 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 CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 12.6 GB, so you would need a larger GPU or CPU offloading.

What is the file size of CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged GGUF?

It depends on the quant: Q2_K is 11.3 GB, Q3_K_S is 13.3 GB, Q3_K_M is 14.7 GB, Q3_K_L is 15.9 GB, IQ4_XS is 16.6 GB, Q4_K_S is 17.5 GB and 5 more.

What context length does CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged support?

The GGUF header lists a trained context length of 40,960 tokens (40K). The KV cache grows with the context you use: about 0.8 GB at 8K and 3.2 GB at 32K.

Where can I download CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged GGUF?

The files are on Hugging Face in the mradermacher/CantoneseLLM-v2.0-30B-A3B-Thinking-Chat-Vector-Merged-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 CantoneseLLM v2.0 30B A3B Thinking Chat Vector Merged use?

Hugging Face lists the license as apache-2.0. 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 CantoneseLLM v2.0 30B on 8 GB VRAM?

Yes, the Q2_K quant requires approximately 12.6 GB memory at 8k context, which fits within an 8 GB VRAM setup with some overhead.

Does this model support vision tasks?

No, the model does not include vision capabilities.

What license applies to this model?

The model is released under the Apache‑2.0 license.

Other GGUFs of CantoneseLLM-v2.0-30B-A3B-Thinking-Chat-Vector-Merged

Similar models

Browse all →

More from mradermacher

All by mradermacher →

More Cantonesellm models

All Cantonesellm →

Other medium models

All medium models →