Carbon 3B needs about 3.2 GB of memory at Q4_K_M (a 2.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 5 quants, from 2.3 GB (Q4_K_M) to 6.9 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.5 GB.
Carbon 3B – a 3.45B‑parameter DNA foundation model
Carbon 3B is a 3.45B‑parameter decoder‑only autoregressive model trained on DNA and RNA sequences. It uses a hybrid 6‑mer tokenizer for DNA and a Qwen3 BPE vocabulary for English text. The native context length is 32,768 6‑mer tokens (~197k DNA base pairs).
Good at
- DNA sequence recovery
- Variant‑effect prediction
- Motif‑perturbation discrimination
Limits
- No vision support
- Requires recent llama.cpp with HybridDNATokenizer
- Not instruction‑tuned
Best for
- Genomic sequence generation
- Variant effect analysis
- DNA‑text joint tasks
Who it’s for. Researchers and developers working on genomic data who need a fast, efficient DNA model. It is suitable for labs with GPU resources and for projects that require hybrid DNA‑text tokenization.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 2.3 GB | 3.2 GB | Fits | Get file → | |
| Q5_K_M | 2.6 GB | 3.5 GB | Fits | Get file → | |
| Q6_K | 3.0 GB | 3.8 GB | Fits | Get file → | |
| Q8_0 | 3.7 GB | 4.6 GB | Fits | Get file → | |
| BF16 | 6.9 GB | 7.8 GB | Tight | Get file → |
For 8 GB VRAM, use Q4_K_M (file size 2.29 GB, ~3.2 GB memory at 8 k context). For 12 GB VRAM, Q5_K_M (2.61 GB, ~3.5 GB). For 16 GB VRAM, Q6_K (2.95 GB, ~3.8 GB).
What GPU do you need for Carbon 3B?
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 Q6_K (3.0 GB file, about 3.8 GB in use).
- 6 GB VRAM: the best quant that fits is Q8_0 (3.7 GB file, about 4.6 GB in use).
- 8 GB VRAM: the best quant that fits is BF16 (6.9 GB file, about 7.8 GB in use).
Which Carbon 3B quant should you choose?
Q4_K_M is the usual starting point for Carbon 3B: it balances file size and quality well. Here it is a 2.3 GB file.
The Q8_0 file is 3.7 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.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Carbon 3B
How much VRAM does Carbon 3B need?
About 3.2 GB at Q4_K_M with an 8K context window. That is the 2.3 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 Carbon 3B on an 8 GB GPU?
Yes, with BF16: the 6.9 GB file needs about 7.8 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of Carbon 3B GGUF?
It depends on the quant: Q4_K_M is 2.3 GB, Q5_K_M is 2.6 GB, Q6_K is 3.0 GB, Q8_0 is 3.7 GB, BF16 is 6.9 GB.
What context length does Carbon 3B support?
The GGUF header lists a trained context length of 32,768 tokens (32K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.5 GB at 32K.
Where can I download Carbon 3B GGUF?
The files are on Hugging Face in the HuggingFaceBio/Carbon-3B-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 Carbon 3B 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 Carbon 3B on 8 GB VRAM?
Yes, the Q4_K_M quantized file uses about 3.2 GB memory at 8 k context, which fits within 8 GB VRAM.
Does Carbon 3B support image inputs?
No, the model does not have vision capabilities.