Pythia 12b needs about 14.1 GB of memory at Q4_K_M (a 7.6 GB file) with an 8K context window, so it fits on a 16 GB GPU. This repository has 21 quants, from 2.6 GB (IQ1_S) to 9.7 GB (Q6_K).
Pythia 12b – 11.8B parameter transformer model in GGUF format
Pythia 12b is a 11.8B parameter GPT-NeoX transformer trained on the Pile. It is released under Apache 2.0 and available in multiple GGUF quantizations. The model supports a 2048 token context window.
Good at
- Large context generation
- Scientific research
- Interpretability studies
Limits
- No vision support
- Requires 8k context for memory estimates
- License is Apache 2.0
Best for
- Research on language model interpretability
- Benchmarking transformer models
- Fine‑tuning on custom datasets
Who it’s for. Researchers and developers needing a large language model for scientific studies. The model is suitable for users who can manage the memory requirements of the various quantizations.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 2.6 GB | 4.6 GB | Fits | Get file → | |
| IQ1_M | 2.8 GB | 4.8 GB | Fits | Get file → | |
| IQ2_XXS | 3.2 GB | 5.2 GB | Fits | Get file → | |
| IQ2_XS | 3.6 GB | 5.6 GB | Fits | Get file → | |
| IQ2_S | 3.8 GB | 5.8 GB | Fits | Get file → | |
| IQ2_M | 4.1 GB | 6.1 GB | Fits | Get file → | |
| Q2_K | 4.5 GB | 6.5 GB | Fits | Get file → | |
| IQ3_XXS | 4.7 GB | 6.7 GB | Fits | Get file → | |
| IQ3_XS | 5.1 GB | 7.1 GB | Fits | Get file → | |
| IQ3_S | 5.2 GB | 7.2 GB | Tight | Get file → | |
| Q3_K_S | 5.2 GB | 7.2 GB | Tight | Get file → | |
| IQ3_M | 5.8 GB | 7.8 GB | Tight | Get file → | |
| Q3_K_M | 6.2 GB | 8.2 GB | Too big | Get file → | |
| IQ4_XS | 6.4 GB | 8.4 GB | Too big | Get file → | |
| Q4_0 | 6.8 GB | 8.8 GB | Too big | Get file → | |
| Q4_K_S | 6.8 GB | 8.8 GB | Too big | Get file → | |
| Q3_K_L | 6.8 GB | 8.8 GB | Too big | Get file → | |
| Q4_K_M | 7.6 GB | 9.6 GB | Too big | Get file → | |
| Q5_K_S | 8.2 GB | 10.2 GB | Too big | Get file → | |
| Q5_K_M | 8.8 GB | 10.8 GB | Too big | Get file → | |
| Q6_K | 9.7 GB | 11.7 GB | Too big | Get file → |
For 8‑GB VRAM, use IQ1_S (2.6 GB file, 9.1 GB memory at 8k context). For 16‑GB VRAM, IQ3_M (5.76 GB file, 12.3 GB memory) or Q4_K_M (7.58 GB file, 14.1 GB memory). For 24‑GB VRAM, Q5_K_M (8.82 GB file, 15.4 GB memory) or Q6_K (9.73 GB file, 16.3 GB memory). Estimates are approximate.
What GPU do you need for Pythia 12b?
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 IQ3_S (5.2 GB file, about 11.7 GB in use).
- 16 GB VRAM: the best quant that fits is Q5_K_M (8.8 GB file, about 15.4 GB in use).
- 24 GB VRAM: the best quant that fits is Q6_K (9.7 GB file, about 16.3 GB in use).
Which Pythia 12b quant should you choose?
Q4_K_M is the usual starting point for Pythia 12b: it balances file size and quality well. Here it is a 7.6 GB file.
If memory is tight, the smallest file is IQ1_S at 2.6 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 Pythia 12b
How much VRAM does Pythia 12b need?
About 14.1 GB at Q4_K_M with an 8K context window. That is the 7.6 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 Pythia 12b on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 9.1 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Pythia 12b GGUF?
It depends on the quant: IQ1_S is 2.6 GB, IQ1_M is 2.8 GB, IQ2_XXS is 3.2 GB, IQ2_XS is 3.6 GB, IQ2_S is 3.8 GB, IQ2_M is 4.1 GB and 15 more.
What context length does Pythia 12b support?
The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 6.0 GB at 8K and 1.5 GB at 2K.
Where can I download Pythia 12b GGUF?
The files are on Hugging Face in the mradermacher/pythia-12b-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 Pythia 12b 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 Pythia 12b on 8 GB VRAM?
Use the IQ1_S quantization; it requires about 9.1 GB memory at 8k context, which is close to 8 GB but may need swapping.
Does Pythia 12b support image inputs?
No, the model has no vision capabilities.
What license applies to Pythia 12b?
The model is released under the Apache 2.0 license.