mradermacher
/

Grafted-Llama2-2x70B-i1-GGUF

@@ -1,6 +1,76 @@
 <!-- ### quantize_version: 2 -->
 <!-- ### output_tensor_quantised: 1 -->
 <!-- ### convert_type: hf -->
 <!-- ### vocab_type:  -->
 <!-- ### tags: nicoboss -->
 weighted/imatrix quants of https://huggingface.co/lodrick-the-lafted/Grafted-Llama2-2x70B

+---
+base_model: lodrick-the-lafted/Grafted-Llama2-2x70B
+language:
+- en
+library_name: transformers
+license: llama2
+quantized_by: mradermacher
+tags:
+- moe
+- merge
+---
+## About
 <!-- ### quantize_version: 2 -->
 <!-- ### output_tensor_quantised: 1 -->
 <!-- ### convert_type: hf -->
 <!-- ### vocab_type:  -->
 <!-- ### tags: nicoboss -->
 weighted/imatrix quants of https://huggingface.co/lodrick-the-lafted/Grafted-Llama2-2x70B
+<!-- provided-files -->
+static quants are available at https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-GGUF
+## Usage
+If you are unsure how to use GGUF files, refer to one of [TheBloke's
+READMEs](https://huggingface.co/TheBloke/KafkaLM-70B-German-V0.1-GGUF) for
+more details, including on how to concatenate multi-part files.
+## Provided Quants
+(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
+| Link | Type | Size/GB | Notes |
+|:-----|:-----|--------:|:------|
+| [GGUF](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ1_S.gguf) | i1-IQ1_S | 26.0 | for the desperate |
+| [GGUF](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ1_M.gguf) | i1-IQ1_M | 28.6 | mostly desperate |
+| [GGUF](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ2_XXS.gguf) | i1-IQ2_XXS | 33.1 |  |
+| [GGUF](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ2_XS.gguf) | i1-IQ2_XS | 36.8 |  |
+| [GGUF](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ2_S.gguf) | i1-IQ2_S | 38.1 |  |
+| [GGUF](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ2_M.gguf) | i1-IQ2_M | 41.6 |  |
+| [GGUF](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q2_K.gguf) | i1-Q2_K | 46.0 | IQ3_XXS probably better |
+| [GGUF](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ3_XXS.gguf) | i1-IQ3_XXS | 48.3 | lower quality |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ3_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ3_XS.gguf.part2of2) | i1-IQ3_XS | 51.3 |  |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ3_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ3_S.gguf.part2of2) | i1-IQ3_S | 54.2 | beats Q3_K* |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q3_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q3_K_S.gguf.part2of2) | i1-Q3_K_S | 54.2 | IQ3_XS probably better |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ3_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ3_M.gguf.part2of2) | i1-IQ3_M | 55.6 |  |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q3_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q3_K_M.gguf.part2of2) | i1-Q3_K_M | 60.2 | IQ3_S probably better |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q3_K_L.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q3_K_L.gguf.part2of2) | i1-Q3_K_L | 65.3 | IQ3_M probably better |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ4_XS.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-IQ4_XS.gguf.part2of2) | i1-IQ4_XS | 66.9 |  |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q4_0.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q4_0.gguf.part2of2) | i1-Q4_0 | 71.0 | fast, low quality |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q4_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q4_K_S.gguf.part2of2) | i1-Q4_K_S | 71.4 | optimal size/speed/quality |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q4_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q4_K_M.gguf.part2of2) | i1-Q4_K_M | 75.7 | fast, recommended |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q5_K_S.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q5_K_S.gguf.part2of2) | i1-Q5_K_S | 86.3 |  |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q5_K_M.gguf.part1of2) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q5_K_M.gguf.part2of2) | i1-Q5_K_M | 88.9 |  |
+| [PART 1](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q6_K.gguf.part1of3) [PART 2](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q6_K.gguf.part2of3) [PART 3](https://huggingface.co/mradermacher/Grafted-Llama2-2x70B-i1-GGUF/resolve/main/Grafted-Llama2-2x70B.i1-Q6_K.gguf.part3of3) | i1-Q6_K | 102.9 | practically like static Q6_K |
+Here is a handy graph by ikawrakow comparing some lower-quality quant
+types (lower is better):
+![image.png](https://www.nethype.de/huggingface_embed/quantpplgraph.png)
+And here are Artefact2's thoughts on the matter:
+https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
+## FAQ / Model Request
+See https://huggingface.co/mradermacher/model_requests for some answers to
+questions you might have and/or if you want some other model quantized.
+## Thanks
+I thank my company, [nethype GmbH](https://www.nethype.de/), for letting
+me use its servers and providing upgrades to my workstation to enable
+this work in my free time. Additional thanks to [@nicoboss](https://huggingface.co/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 -->