Quantization made by Richard Erkhov. [Github](https://github.com/RichardErkhov) [Discord](https://discord.gg/pvy7H8DZMG) [Request more models](https://github.com/RichardErkhov/quant_request) Llama-3-70b-Arimas-story-RP-V2.1 - GGUF - Model creator: https://huggingface.co/ryzen88/ - Original model: https://huggingface.co/ryzen88/Llama-3-70b-Arimas-story-RP-V2.1/ | Name | Quant method | Size | | ---- | ---- | ---- | | [Llama-3-70b-Arimas-story-RP-V2.1.Q2_K.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/blob/main/Llama-3-70b-Arimas-story-RP-V2.1.Q2_K.gguf) | Q2_K | 24.56GB | | [Llama-3-70b-Arimas-story-RP-V2.1.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/blob/main/Llama-3-70b-Arimas-story-RP-V2.1.IQ3_XS.gguf) | IQ3_XS | 27.29GB | | [Llama-3-70b-Arimas-story-RP-V2.1.IQ3_S.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/blob/main/Llama-3-70b-Arimas-story-RP-V2.1.IQ3_S.gguf) | IQ3_S | 28.79GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/blob/main/Llama-3-70b-Arimas-story-RP-V2.1.Q3_K_S.gguf) | Q3_K_S | 28.79GB | | [Llama-3-70b-Arimas-story-RP-V2.1.IQ3_M.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/blob/main/Llama-3-70b-Arimas-story-RP-V2.1.IQ3_M.gguf) | IQ3_M | 29.74GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q3_K.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/blob/main/Llama-3-70b-Arimas-story-RP-V2.1.Q3_K.gguf) | Q3_K | 31.91GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/blob/main/Llama-3-70b-Arimas-story-RP-V2.1.Q3_K_M.gguf) | Q3_K_M | 31.91GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/blob/main/Llama-3-70b-Arimas-story-RP-V2.1.Q3_K_L.gguf) | Q3_K_L | 34.59GB | | [Llama-3-70b-Arimas-story-RP-V2.1.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/blob/main/Llama-3-70b-Arimas-story-RP-V2.1.IQ4_XS.gguf) | IQ4_XS | 35.64GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q4_0.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/blob/main/Llama-3-70b-Arimas-story-RP-V2.1.Q4_0.gguf) | Q4_0 | 37.22GB | | [Llama-3-70b-Arimas-story-RP-V2.1.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/tree/main/) | IQ4_NL | 37.58GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/tree/main/) | Q4_K_S | 37.58GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q4_K.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/tree/main/) | Q4_K | 39.6GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/tree/main/) | Q4_K_M | 39.6GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q4_1.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/tree/main/) | Q4_1 | 41.27GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q5_0.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/tree/main/) | Q5_0 | 45.32GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/tree/main/) | Q5_K_S | 45.32GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q5_K.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/tree/main/) | Q5_K | 46.52GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/tree/main/) | Q5_K_M | 46.52GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q5_1.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/tree/main/) | Q5_1 | 49.36GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q6_K.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/tree/main/) | Q6_K | 53.91GB | | [Llama-3-70b-Arimas-story-RP-V2.1.Q8_0.gguf](https://huggingface.co/RichardErkhov/ryzen88_-_Llama-3-70b-Arimas-story-RP-V2.1-gguf/tree/main/) | Q8_0 | 69.83GB | Original model description: --- base_model: [] library_name: transformers tags: - mergekit - merge - llama 3 - Model stock --- # Merge_XL_model_Stock Ofcourse the model is still fully focussed on uncensored long context Roleplay and Story. By far the best itteration. This model switches to the Smaug instruct 32K for the base bodel. Expanded with Giraffe and Gradient to keep a robuust long context window. Higgs and cat for most of the story and RP aspects. Hermes and Chinese chat are for overall intelligence and understanding. ## Merge Details ### Merge Method This model was merged using the [Model Stock](https://arxiv.org/abs/2403.19522) merge method using \Smaug-Llama-3-70B-Instruct-32K as a base. ### Models Merged The following models were included in the merge: * \Llama-3-Giraffe-70B-Instruct * \Llama-3-70B-Instruct-Gradient-262k * \Hermes-2-Theta-Llama-3-70B * \Higgs-Llama-3-70B * \Llama3-70B-Chinese-Chat * \Meta-LLama-3-Cat-A-LLama-70b ### Configuration The following YAML configuration was used to produce this model: ```yaml models: - model: \Smaug-Llama-3-70B-Instruct-32K - model: \Llama-3-70B-Instruct-Gradient-262k - model: \Llama-3-Giraffe-70B-Instruct - model: \Higgs-Llama-3-70B - model: \Llama3-70B-Chinese-Chat - model: \Meta-LLama-3-Cat-A-LLama-70b - model: \Hermes-2-Theta-Llama-3-70B merge_method: model_stock base_model: \Smaug-Llama-3-70B-Instruct-32K dtype: bfloat16 ``` Any suggestions are very welcome My personal sampling settings are: "temp": 1, "temperature_last": true, "top_p": 1, "top_k": 0, "top_a": 0, "tfs": 1, "typical_p": 1, "min_p": 0.05, "rep_pen": 1.05, "rep_pen_range": 4096, "rep_pen_decay": 0, "rep_pen_slope": 1,