Downloads · 30 days
2K
2% of all-time downloads
mradermacher/Llama-3.3-70B-Instruct-heretic-i1-GGUF
Llama-3.3-70B-Instruct-heretic-i1-GGUF is a machine learning model from mradermacher. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as llama3.3.
weighted/imatrix quants of https://huggingface.co/kldzj/Llama-3.3-70B-Instruct-heretic
Downloads · 30 days
2K
2% of all-time downloads
All-time downloads
97.4K
Public
Repo size
1.5 TB
Likes
0
Public
Click a slice to open those files.
.gguf1.4 TB · 96%
From the Hugging Face model README
weighted/imatrix quants of https://huggingface.co/kldzj/Llama-3.3-70B-Instruct-heretic
<!-- provided-files -->For a convenient overview and download list, visit our model page for this model.
static quants are available at https://huggingface.co/mradermacher/Llama-3.3-70B-Instruct-heretic-GGUF
If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files.
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|---|---|---|---|
| PART 1 PART 2 | imatrix | 0.1 | imatrix file (for creating your own quants) |
| PART 1 PART 2 | i1-IQ1_S | 30.8 | for the desperate |
| PART 1 PART 2 | i1-IQ1_M | 33.6 | mostly desperate |
| PART 1 PART 2 | i1-IQ2_XXS | 38.3 | |
| PART 1 PART 2 | i1-IQ2_XS | 42.4 | |
| PART 1 PART 2 | i1-IQ2_S | 44.6 | |
| PART 1 PART 2 | i1-IQ2_M | 48.3 | |
| PART 1 PART 2 | i1-Q2_K_S | 49.0 | very low quality |
| PART 1 PART 2 | i1-Q2_K | 52.9 | IQ3_XXS probably better |
| PART 1 PART 2 | i1-IQ3_XXS | 55.0 | lower quality |
| PART 1 PART 2 | i1-IQ3_XS | 58.7 | |
| PART 1 PART 2 | i1-IQ3_S | 61.9 | beats Q3_K* |
| PART 1 PART 2 | i1-Q3_K_S | 61.9 | IQ3_XS probably better |
| PART 1 PART 2 | i1-IQ3_M | 64.0 | |
| PART 1 PART 2 | i1-Q3_K_M | 68.6 | IQ3_S probably better |
| PART 1 PART 2 | i1-Q3_K_L | 74.4 | IQ3_M probably better |
| PART 1 PART 2 | i1-IQ4_XS | 75.9 | |
| PART 1 PART 2 | i1-Q4_0 | 80.3 | fast, low quality |
| PART 1 PART 2 | i1-Q4_K_S | 80.8 | optimal size/speed/quality |
| PART 1 PART 2 | i1-Q4_K_M | 85.1 | fast, recommended |
| PART 1 PART 2 | i1-Q4_1 | 88.7 | |
| PART 1 PART 2 | i1-Q5_K_S | 97.4 | |
| PART 1 PART 2 | i1-Q5_K_M | 100.0 | |
| PART 1 PART 2 PART 3 | i1-Q6_K | 115.9 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):

And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized.
I thank my company, nethype GmbH, for letting me use its servers and providing upgrades to my workstation to enable this work in my free time. Additional thanks to @nicoboss for giving me access to his private supercomputer, enabling me to provide many more imatrix quants, at much higher quality, than I would otherwise be able to.
<!-- end -->