Carbon-500M GGUF
Available files
| Quant | File | Size |
|---|---|---|
| BF16 | carbon-500m-bf16.gguf |
982.2 MB |
| Q8_0 | carbon-500m-q8_0.gguf |
524.6 MB |
| Q6_K | carbon-500m-q6_k.gguf |
443.2 MB |
| Q5_K_M | carbon-500m-q5_k_m.gguf |
405.1 MB |
| Q4_K_M | carbon-500m-q4_k_m.gguf |
369.2 MB |
K-quants are produced with --token-embedding-type Q8_0 --output-tensor-type Q8_0
so the sensitive embedding/output tensors stay at Q8_0 precision (important for
this model's tight DNA-vs-text logit margins).
How to use
Click Use this model > llama.cpp on this page for a ready-made snippet. To install llama.cpp, follow llama.app or download a prebuilt binary from the releases page.
Carbon-500M is a base DNA model, not a chat model, so use text completion rather than the chat UI. Start the prompt
with <dna>:
llama completion -hf HuggingFaceBio/Carbon-500M-GGUF:Q8_0 -p '<dna>ATGCGCTAGCTACGATCGATCGTAGCTAGCTAGCTAGCTACG' -n 128 --temp 0
llama serve -hf HuggingFaceBio/Carbon-500M-GGUF:Q8_0 serves the same model over HTTP; send raw prompts to its /completion endpoint.
Carbon-500M shares its vocabulary with Carbon-3B and Carbon-8B, so it works as their draft model for speculative
decoding: add -hfd HuggingFaceBio/Carbon-500M-GGUF:Q8_0 when running either of them.
See also
- Source weights: HuggingFaceBio/Carbon-500M
- Other GGUF variants: 500M · 3B · 8B
License
Apache-2.0, inherited from the source model.