From 7465f519b9c93363a87d40eb159d5ab14d9e0ee2 Mon Sep 17 00:00:00 2001 From: Bartowski Date: Wed, 12 Mar 2025 07:40:42 +0000 Subject: [PATCH] Upload README.md with huggingface_hub --- README.md | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index 9cb0ad6..6c5a8ac 100644 --- a/README.md +++ b/README.md @@ -1,8 +1,6 @@ --- quantized_by: bartowski pipeline_tag: text-generation -license: apache-2.0 -base_model: RekaAI/reka-flash-3 --- ## Llamacpp imatrix Quantizations of reka-flash-3 by RekaAI @@ -20,9 +18,13 @@ Run them directly with [llama.cpp](https://github.com/ggerganov/llama.cpp), or a ## Prompt format ``` -human: {prompt} assistant: +human: {system_prompt} {prompt} assistant: ``` +## What's new: + +Fix chat template + ## Download a file (not the whole branch) from below: | Filename | Quant type | File Size | Split | Description | @@ -52,6 +54,8 @@ human: {prompt} assistant: | [reka-flash-3-Q2_K.gguf](https://huggingface.co/bartowski/RekaAI_reka-flash-3-GGUF/blob/main/RekaAI_reka-flash-3-Q2_K.gguf) | Q2_K | 8.63GB | false | Very low quality but surprisingly usable. | | [reka-flash-3-IQ2_M.gguf](https://huggingface.co/bartowski/RekaAI_reka-flash-3-GGUF/blob/main/RekaAI_reka-flash-3-IQ2_M.gguf) | IQ2_M | 8.51GB | false | Relatively low quality, uses SOTA techniques to be surprisingly usable. | | [reka-flash-3-IQ2_S.gguf](https://huggingface.co/bartowski/RekaAI_reka-flash-3-GGUF/blob/main/RekaAI_reka-flash-3-IQ2_S.gguf) | IQ2_S | 8.12GB | false | Low quality, uses SOTA techniques to be usable. | +| [reka-flash-3-IQ2_XS.gguf](https://huggingface.co/bartowski/RekaAI_reka-flash-3-GGUF/blob/main/RekaAI_reka-flash-3-IQ2_XS.gguf) | IQ2_XS | 7.83GB | false | Low quality, uses SOTA techniques to be usable. | +| [reka-flash-3-IQ2_XXS.gguf](https://huggingface.co/bartowski/RekaAI_reka-flash-3-GGUF/blob/main/RekaAI_reka-flash-3-IQ2_XXS.gguf) | IQ2_XXS | 7.39GB | false | Very low quality, uses SOTA techniques to be usable. | ## Embed/output weights