From d444c2fbcf337b3f95fa2f3b94d4d710ac9ed346 Mon Sep 17 00:00:00 2001 From: "LocalAI [bot]" <139863280+localai-bot@users.noreply.github.com> Date: Fri, 27 Sep 2024 08:54:36 +0200 Subject: [PATCH] chore(model-gallery): :arrow_up: update checksum (#3675) :arrow_up: Checksum updates in gallery/index.yaml Signed-off-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com> Co-authored-by: mudler <2420543+mudler@users.noreply.github.com> Signed-off-by: Siddharth More --- gallery/index.yaml | 34 +++++++++++++++++----------------- 1 file changed, 17 insertions(+), 17 deletions(-) diff --git a/gallery/index.yaml b/gallery/index.yaml index de38c3d5881..4b668061b12 100644 --- a/gallery/index.yaml +++ b/gallery/index.yaml @@ -59,8 +59,8 @@ - filename: llama-3.2-1b-instruct-q8_0.gguf sha256: ba345c83bf5cc679c653b853c46517eea5a34f03ed2205449db77184d9ae62a9 uri: huggingface://hugging-quants/Llama-3.2-1B-Instruct-Q8_0-GGUF/llama-3.2-1b-instruct-q8_0.gguf -## Qwen2.5 - &qwen25 + ## Qwen2.5 name: "qwen2.5-14b-instruct" url: "github:mudler/LocalAI/gallery/chatml.yaml@master" license: apache-2.0 @@ -89,11 +89,11 @@ - https://huggingface.co/bartowski/Qwen2.5-Math-7B-Instruct-GGUF - https://huggingface.co/Qwen/Qwen2.5-Math-7B-Instruct description: | - In August 2024, we released the first series of mathematical LLMs - Qwen2-Math - of our Qwen family. A month later, we have upgraded it and open-sourced Qwen2.5-Math series, including base models Qwen2.5-Math-1.5B/7B/72B, instruction-tuned models Qwen2.5-Math-1.5B/7B/72B-Instruct, and mathematical reward model Qwen2.5-Math-RM-72B. + In August 2024, we released the first series of mathematical LLMs - Qwen2-Math - of our Qwen family. A month later, we have upgraded it and open-sourced Qwen2.5-Math series, including base models Qwen2.5-Math-1.5B/7B/72B, instruction-tuned models Qwen2.5-Math-1.5B/7B/72B-Instruct, and mathematical reward model Qwen2.5-Math-RM-72B. - Unlike Qwen2-Math series which only supports using Chain-of-Thught (CoT) to solve English math problems, Qwen2.5-Math series is expanded to support using both CoT and Tool-integrated Reasoning (TIR) to solve math problems in both Chinese and English. The Qwen2.5-Math series models have achieved significant performance improvements compared to the Qwen2-Math series models on the Chinese and English mathematics benchmarks with CoT. + Unlike Qwen2-Math series which only supports using Chain-of-Thught (CoT) to solve English math problems, Qwen2.5-Math series is expanded to support using both CoT and Tool-integrated Reasoning (TIR) to solve math problems in both Chinese and English. The Qwen2.5-Math series models have achieved significant performance improvements compared to the Qwen2-Math series models on the Chinese and English mathematics benchmarks with CoT. - The base models of Qwen2-Math are initialized with Qwen2-1.5B/7B/72B, and then pretrained on a meticulously designed Mathematics-specific Corpus. This corpus contains large-scale high-quality mathematical web texts, books, codes, exam questions, and mathematical pre-training data synthesized by Qwen2. + The base models of Qwen2-Math are initialized with Qwen2-1.5B/7B/72B, and then pretrained on a meticulously designed Mathematics-specific Corpus. This corpus contains large-scale high-quality mathematical web texts, books, codes, exam questions, and mathematical pre-training data synthesized by Qwen2. overrides: parameters: model: Qwen2.5-Math-7B-Instruct-Q4_K_M.gguf @@ -195,8 +195,8 @@ model: Qwen2.5-32B.Q4_K_M.gguf files: - filename: Qwen2.5-32B.Q4_K_M.gguf - sha256: 02703e27c8b964db445444581a6937ad7538f0c32a100b26b49fa0e8ff527155 uri: huggingface://mradermacher/Qwen2.5-32B-GGUF/Qwen2.5-32B.Q4_K_M.gguf + sha256: fa42a4067e3630929202b6bb1ef5cebc43c1898494aedfd567b7d53c7a9d84a6 - !!merge <<: *qwen25 name: "qwen2.5-32b-instruct" urls: @@ -221,8 +221,8 @@ - filename: Qwen2.5-72B-Instruct-Q4_K_M.gguf sha256: e4c8fad16946be8cf0bbf67eb8f4e18fc7415a5a6d2854b4cda453edb4082545 uri: huggingface://bartowski/Qwen2.5-72B-Instruct-GGUF/Qwen2.5-72B-Instruct-Q4_K_M.gguf -## SmolLM - &smollm + ## SmolLM url: "github:mudler/LocalAI/gallery/chatml.yaml@master" name: "smollm-1.7b-instruct" icon: https://huggingface.co/datasets/HuggingFaceTB/images/resolve/main/banner_smol.png @@ -651,9 +651,9 @@ - https://huggingface.co/leafspark/Reflection-Llama-3.1-70B-bf16 - https://huggingface.co/senseable/Reflection-Llama-3.1-70B-gguf description: | - Reflection Llama-3.1 70B is (currently) the world's top open-source LLM, trained with a new technique called Reflection-Tuning that teaches a LLM to detect mistakes in its reasoning and correct course. + Reflection Llama-3.1 70B is (currently) the world's top open-source LLM, trained with a new technique called Reflection-Tuning that teaches a LLM to detect mistakes in its reasoning and correct course. - The model was trained on synthetic data generated by Glaive. If you're training a model, Glaive is incredible — use them. + The model was trained on synthetic data generated by Glaive. If you're training a model, Glaive is incredible — use them. overrides: parameters: model: Reflection-Llama-3.1-70B-q4_k_m.gguf @@ -973,15 +973,15 @@ - https://huggingface.co/Sao10K/L3.1-8B-Niitama-v1.1 - https://huggingface.co/Lewdiculous/L3.1-8B-Niitama-v1.1-GGUF-IQ-Imatrix description: | - GGUF-IQ-Imatrix quants for Sao10K/L3.1-8B-Niitama-v1.1 - Here's the subjectively superior L3 version: L3-8B-Niitama-v1 - An experimental model using experimental methods. + GGUF-IQ-Imatrix quants for Sao10K/L3.1-8B-Niitama-v1.1 + Here's the subjectively superior L3 version: L3-8B-Niitama-v1 + An experimental model using experimental methods. - More detail on it: + More detail on it: - Tamamo and Niitama are made from the same data. Literally. The only thing that's changed is how theyre shuffled and formatted. Yet, I get wildly different results. + Tamamo and Niitama are made from the same data. Literally. The only thing that's changed is how theyre shuffled and formatted. Yet, I get wildly different results. - Interesting, eh? Feels kinda not as good compared to the l3 version, but it's aight. + Interesting, eh? Feels kinda not as good compared to the l3 version, but it's aight. overrides: parameters: model: L3.1-8B-Niitama-v1.1-Q4_K_M-imat.gguf @@ -1606,8 +1606,8 @@ urls: - https://huggingface.co/Lewdiculous/MN-12B-Lyra-v4-GGUF-IQ-Imatrix description: | - A finetune of Mistral Nemo by Sao10K. - Uses the ChatML prompt format. + A finetune of Mistral Nemo by Sao10K. + Uses the ChatML prompt format. overrides: parameters: model: MN-12B-Lyra-v4-Q4_K_M-imat.gguf @@ -2134,7 +2134,7 @@ - https://huggingface.co/EpistemeAI/Athena-codegemma-2-2b-it - https://huggingface.co/mradermacher/Athena-codegemma-2-2b-it-GGUF description: | - Supervised fine tuned (sft unsloth) for coding with EpistemeAI coding dataset. + Supervised fine tuned (sft unsloth) for coding with EpistemeAI coding dataset. overrides: parameters: model: Athena-codegemma-2-2b-it.Q4_K_M.gguf