HarmBench Llama 2 13b Cls needs about 15.1 GB of memory at Q4_K_M (a 7.9 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 25 quants, from 2.9 GB (IQ1_S) to 10.7 GB (Q6_K).
HarmBench Llama 2 13b Cls – a text behavior classifier
HarmBench Llama 2 13b Cls is a 13 billion parameter classifier built on Llama 2. It is trained on distilled GPT‑4 data and supports standard and contextual text behaviors. The model is released under the MIT license.
Good at
- Classifying text behaviors
- Distilled GPT‑4 training
- MIT licensed
Limits
- No vision support
- Context limited to 2048 tokens
- No mixture of experts
Best for
- Content moderation
- Safety filtering
- Behavior analysis
Who it’s for. Developers needing a lightweight classifier for text safety. Researchers working on harm detection can use the model for quick inference.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 2.9 GB | 5.1 GB | Fits | Get file → | |
| IQ1_M | 3.1 GB | 5.3 GB | Fits | Get file → | |
| IQ2_XXS | 3.5 GB | 5.7 GB | Fits | Get file → | |
| IQ2_XS | 3.9 GB | 6.1 GB | Fits | Get file → | |
| IQ2_S | 4.2 GB | 6.4 GB | Fits | Get file → | |
| Q2_K_S | 4.4 GB | 6.6 GB | Fits | Get file → | |
| IQ2_M | 4.5 GB | 6.7 GB | Fits | Get file → | |
| Q2_K | 4.9 GB | 7.0 GB | Fits | Get file → | |
| IQ3_XXS | 5.0 GB | 7.1 GB | Fits | Get file → | |
| IQ3_XS | 5.4 GB | 7.5 GB | Tight | Get file → | |
| IQ3_S | 5.7 GB | 7.8 GB | Tight | Get file → | |
| Q3_K_S | 5.7 GB | 7.8 GB | Tight | Get file → | |
| IQ3_M | 6.0 GB | 8.2 GB | Too big | Get file → | |
| Q3_K_M | 6.3 GB | 8.5 GB | Too big | Get file → | |
| Q3_K_L | 6.9 GB | 9.1 GB | Too big | Get file → | |
| IQ4_XS | 7.0 GB | 9.1 GB | Too big | Get file → | |
| Q4_0 | 7.4 GB | 9.5 GB | Too big | Get file → | |
| Q4_0 | 7.4 GB | 9.5 GB | Too big | Get file → | |
| Q4_0 | 7.4 GB | 9.5 GB | Too big | Get file → | |
| Q4_0 | 7.4 GB | 9.6 GB | Too big | Get file → | |
| Q4_K_S | 7.4 GB | 9.6 GB | Too big | Get file → | |
| Q4_K_M | 7.9 GB | 10.0 GB | Too big | Get file → | |
| Q5_K_S | 9.0 GB | 11.2 GB | Too big | Get file → | |
| Q5_K_M | 9.2 GB | 11.4 GB | Too big | Get file → | |
| Q6_K | 10.7 GB | 12.9 GB | Too big | Get file → |
For 8 GB VRAM, use IQ1_S (2.9 GB file, 10.1 GB memory at 8k context). With 12 GB VRAM, IQ2_XS (3.89 GB file, 11.1 GB memory) or Q2_K (4.85 GB file, 12.1 GB memory) are suitable. For 16 GB VRAM, Q3_K_M (6.34 GB file, 13.5 GB memory) or Q4_K_M (7.87 GB file, 15.1 GB memory) can be used.
What GPU do you need for HarmBench Llama 2 13b Cls?
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: the best quant that fits is IQ2_M (4.5 GB file, about 11.7 GB in use).
- 16 GB VRAM: the best quant that fits is Q4_K_M (7.9 GB file, about 15.1 GB in use).
- 24 GB VRAM: the best quant that fits is Q6_K (10.7 GB file, about 17.9 GB in use).
Which HarmBench Llama 2 13b Cls quant should you choose?
Q4_K_M is the usual starting point for HarmBench Llama 2 13b Cls: it balances file size and quality well. Here it is a 7.9 GB file.
If memory is tight, the smallest file is IQ1_S at 2.9 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 HarmBench Llama 2 13b Cls
How much VRAM does HarmBench Llama 2 13b Cls need?
About 15.1 GB at Q4_K_M with an 8K context window. That is the 7.9 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 HarmBench Llama 2 13b Cls on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 10.1 GB, so you would need a larger GPU or CPU offloading.
What is the file size of HarmBench Llama 2 13b Cls GGUF?
It depends on the quant: IQ1_S is 2.9 GB, IQ1_M is 3.1 GB, IQ2_XXS is 3.5 GB, IQ2_XS is 3.9 GB, IQ2_S is 4.2 GB, Q2_K_S is 4.4 GB and 19 more.
What context length does HarmBench Llama 2 13b Cls support?
The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 6.7 GB at 8K and 1.7 GB at 2K.
Where can I download HarmBench Llama 2 13b Cls GGUF?
The files are on Hugging Face in the mradermacher/HarmBench-Llama-2-13b-cls-i1-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 HarmBench Llama 2 13b Cls use?
Hugging Face lists the license as mit. 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 IQ1_S quant requires 10.1 GB memory at 8k context, which fits within 8 GB VRAM with some headroom.
Does this model support image inputs?
No, it only handles text inputs and has no vision capabilities.
What license does the model use?
The model is released under the MIT license.