--- base_model: Snowflake/snowflake-arctic-instruct language: - en library_name: transformers license: apache-2.0 quantized_by: mradermacher tags: - snowflake - arctic - moe --- ## About weighted/imatrix quants of https://huggingface.co/Snowflake/snowflake-arctic-instruct static quants are available at https://huggingface.co/mradermacher/snowflake-arctic-instruct-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 | |:-----|:-----|--------:|:------| | [PART 1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q2_K.gguf.part1of4) [PART 2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q2_K.gguf.part2of4) [PART 3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q2_K.gguf.part3of4) [PART 4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q2_K.gguf.part4of4) | i1-Q2_K | 173.7 | IQ3_XXS probably better | | [PART 1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_XXS.gguf.part1of4) [PART 2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_XXS.gguf.part2of4) [PART 3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_XXS.gguf.part3of4) [PART 4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_XXS.gguf.part4of4) | i1-IQ3_XXS | 183.5 | lower quality | | [P1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_S.gguf.part1of5) [P2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_S.gguf.part2of5) [P3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_S.gguf.part3of5) [P4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_S.gguf.part4of5) [P5](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_S.gguf.part5of5) | i1-Q3_K_S | 205.9 | IQ3_XS probably better | | [P1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_S.gguf.part1of5) [P2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_S.gguf.part2of5) [P3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_S.gguf.part3of5) [P4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_S.gguf.part4of5) [P5](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_S.gguf.part5of5) | i1-IQ3_S | 206.0 | beats Q3_K* | | [P1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_M.gguf.part1of5) [P2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_M.gguf.part2of5) [P3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_M.gguf.part3of5) [P4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_M.gguf.part4of5) [P5](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ3_M.gguf.part5of5) | i1-IQ3_M | 208.6 | | | [P1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_M.gguf.part1of5) [P2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_M.gguf.part2of5) [P3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_M.gguf.part3of5) [P4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_M.gguf.part4of5) [P5](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_M.gguf.part5of5) | i1-Q3_K_M | 228.3 | IQ3_S probably better | | [P1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_L.gguf.part1of6) [P2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_L.gguf.part2of6) [P3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_L.gguf.part3of6) [P4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_L.gguf.part4of6) [P5](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_L.gguf.part5of6) [P6](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q3_K_L.gguf.part6of6) | i1-Q3_K_L | 247.2 | IQ3_M probably better | | [P1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ4_XS.gguf.part1of6) [P2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ4_XS.gguf.part2of6) [P3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ4_XS.gguf.part3of6) [P4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ4_XS.gguf.part4of6) [P5](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ4_XS.gguf.part5of6) [P6](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-IQ4_XS.gguf.part6of6) | i1-IQ4_XS | 254.6 | | | [P1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_0.gguf.part1of6) [P2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_0.gguf.part2of6) [P3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_0.gguf.part3of6) [P4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_0.gguf.part4of6) [P5](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_0.gguf.part5of6) [P6](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_0.gguf.part6of6) | i1-Q4_0 | 270.6 | fast, low quality | | [P1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_K_S.gguf.part1of6) [P2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_K_S.gguf.part2of6) [P3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_K_S.gguf.part3of6) [P4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_K_S.gguf.part4of6) [P5](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_K_S.gguf.part5of6) [P6](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_K_S.gguf.part6of6) | i1-Q4_K_S | 271.7 | optimal size/speed/quality | | [P1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_K_M.gguf.part1of6) [P2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_K_M.gguf.part2of6) [P3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_K_M.gguf.part3of6) [P4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_K_M.gguf.part4of6) [P5](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_K_M.gguf.part5of6) [P6](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q4_K_M.gguf.part6of6) | i1-Q4_K_M | 289.3 | fast, recommended | | [P1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_S.gguf.part1of7) [P2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_S.gguf.part2of7) [P3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_S.gguf.part3of7) [P4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_S.gguf.part4of7) [P5](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_S.gguf.part5of7) [P6](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_S.gguf.part6of7) [P7](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_S.gguf.part7of7) | i1-Q5_K_S | 329.3 | | | [P1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_M.gguf.part1of7) [P2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_M.gguf.part2of7) [P3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_M.gguf.part3of7) [P4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_M.gguf.part4of7) [P5](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_M.gguf.part5of7) [P6](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_M.gguf.part6of7) [P7](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q5_K_M.gguf.part7of7) | i1-Q5_K_M | 339.5 | | | [P1](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q6_K.gguf.part1of8) [P2](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q6_K.gguf.part2of8) [P3](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q6_K.gguf.part3of8) [P4](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q6_K.gguf.part4of8) [P5](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q6_K.gguf.part5of8) [P6](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q6_K.gguf.part6of8) [P7](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q6_K.gguf.part7of8) [P8](https://huggingface.co/mradermacher/snowflake-arctic-instruct-i1-GGUF/resolve/main/snowflake-arctic-instruct.i1-Q6_K.gguf.part8of8) | i1-Q6_K | 392.8 | 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.