diff --git a/models/capybara-34b/model.json b/models/capybara-34b/model.json index 74eb27eb9..e5834a66d 100644 --- a/models/capybara-34b/model.json +++ b/models/capybara-34b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/Nous-Capybara-34B-GGUF/resolve/main/nous-capybara-34b.Q5_K_M.gguf", "id": "capybara-34b", "object": "model", - "name": "Capybara 200k 34B", + "name": "Capybara 200k 34B Q5", "version": "1.0", "description": "Nous Capybara 34B, a variant of the Yi-34B model, is the first Nous model with a 200K context length, trained for three epochs on the innovative Capybara dataset.", "format": "gguf", diff --git a/models/deepseek-coder-1.3b/model.json b/models/deepseek-coder-1.3b/model.json index dac90423d..9d22f9558 100644 --- a/models/deepseek-coder-1.3b/model.json +++ b/models/deepseek-coder-1.3b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/deepseek-coder-1.3b-base-GGUF/resolve/main/deepseek-coder-1.3b-base.Q4_K_M.gguf", "id": "deepseek-coder-1.3b", "object": "model", - "name": "Deepseek Coder 1.3B", + "name": "Deepseek Coder 1.3B Q4", "version": "1.0", "description": "", "format": "gguf", @@ -16,7 +16,7 @@ "max_tokens": 2048 }, "metadata": { - "author": "deepseek, The Bloke", + "author": "Deepseek, The Bloke", "tags": ["Community Recommended", "Code", "Small size"], "size": 870000000 } diff --git a/models/llama2-chat-70b-q4/model.json b/models/llama2-chat-70b-q4/model.json index 39c0f6474..4ab3ac555 100644 --- a/models/llama2-chat-70b-q4/model.json +++ b/models/llama2-chat-70b-q4/model.json @@ -18,7 +18,7 @@ "metadata": { "author": "MetaAI, The Bloke", "tags": ["Foundational Model", "General", "Code"], - "size": 4080000000 + "size": 43920000000 } } \ No newline at end of file diff --git a/models/llama2-chat-7b-q4/model.json b/models/llama2-chat-7b-q4/model.json index be302d144..7b52e1767 100644 --- a/models/llama2-chat-7b-q4/model.json +++ b/models/llama2-chat-7b-q4/model.json @@ -4,7 +4,7 @@ "object": "model", "name": "Llama 2 Chat 7B Q4", "version": "1.0", - "description": "This is a 4-bit quantized version of Meta AI's Llama 2 Chat 7b model.", + "description": "This is a 4-bit quantized iteration of Meta AI's Llama 2 Chat 7b model, specifically designed for a comprehensive understanding through training on extensive internet data.", "format": "gguf", "settings": { "ctx_len": 2048, diff --git a/models/llama2-chat-7b-q5/model.json b/models/llama2-chat-7b-q5/model.json index 8a93327d6..b19213e46 100644 --- a/models/llama2-chat-7b-q5/model.json +++ b/models/llama2-chat-7b-q5/model.json @@ -4,7 +4,7 @@ "object": "model", "name": "Llama 2 Chat 7B Q5", "version": "1.0", - "description": "This is a 5-bit quantized version of Meta AI's Llama 2 Chat 7b model.", + "description": "This is a 5-bit quantized iteration of Meta AI's Llama 2 Chat 7b model, specifically designed for a comprehensive understanding through training on extensive internet data.", "format": "gguf", "settings": { "ctx_len": 2048, diff --git a/models/lzlv-70b/model.json b/models/lzlv-70b/model.json index 844bdadf4..e4247eb88 100644 --- a/models/lzlv-70b/model.json +++ b/models/lzlv-70b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/lzlv_70B-GGUF/resolve/main/lzlv_70b_fp16_hf.Q5_K_M.gguf", "id": "lzlv-70b", "object": "model", - "name": "Lzlv 70B", + "name": "Lzlv 70B Q4", "version": "1.0", "description": "lzlv_70B is a sophisticated AI model designed for roleplaying and creative tasks. This merge aims to combine intelligence with creativity, seemingly outperforming its individual components in complex scenarios and creative outputs.", "format": "gguf", diff --git a/models/mistral-ins-7b-q4/model.json b/models/mistral-ins-7b-q4/model.json index 51b5c231e..dd3360700 100644 --- a/models/mistral-ins-7b-q4/model.json +++ b/models/mistral-ins-7b-q4/model.json @@ -4,7 +4,7 @@ "object": "model", "name": "Mistral Instruct 7B Q4", "version": "1.0", - "description": "This is a 4-bit quantized version of MistralAI's Mistral Instruct 7B model.", + "description": "This is a 4-bit quantized iteration of MistralAI's Mistral Instruct 7b model, specifically designed for a comprehensive understanding through training on extensive internet data.", "format": "gguf", "settings": { "ctx_len": 2048, diff --git a/models/mistral-ins-7b-q5/model.json b/models/mistral-ins-7b-q5/model.json index 46c5f5378..5b86ab538 100644 --- a/models/mistral-ins-7b-q5/model.json +++ b/models/mistral-ins-7b-q5/model.json @@ -4,7 +4,7 @@ "object": "model", "name": "Mistral Instruct 7B Q5", "version": "1.0", - "description": "This is a 5-bit quantized version of MistralAI's Mistral Instruct 7B model.", + "description": "This is a 5-bit quantized iteration of MistralAI's Mistral Instruct 7b model, specifically designed for a comprehensive understanding through training on extensive internet data.", "format": "gguf", "settings": { "ctx_len": 2048, diff --git a/models/neural-chat-7b/model.json b/models/neural-chat-7b/model.json index cac717b8a..234a2ff80 100644 --- a/models/neural-chat-7b/model.json +++ b/models/neural-chat-7b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/neural-chat-7B-v3-1-GGUF/resolve/main/neural-chat-7b-v3-1.Q4_K_M.gguf", "id": "neural-chat-7b", "object": "model", - "name": "Neural Chat 7B", + "name": "Neural Chat 7B Q4", "version": "1.0", "description": "The Neural Chat 7B model, developed on the foundation of mistralai/Mistral-7B-v0.1, has been fine-tuned using the Open-Orca/SlimOrca dataset and aligned with the Direct Preference Optimization (DPO) algorithm. It has demonstrated substantial improvements in various AI tasks and performance well on the open_llm_leaderboard.", "format": "gguf", diff --git a/models/noromaid-20b/model.json b/models/noromaid-20b/model.json index 61870e855..da69cba3e 100644 --- a/models/noromaid-20b/model.json +++ b/models/noromaid-20b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/Noromaid-20B-v0.1.1-GGUF/resolve/main/noromaid-20b-v0.1.1.Q4_K_M.gguf", "id": "noromaid-20b", "object": "model", - "name": "Noromaid 20B", + "name": "Noromaid 20B Q4", "version": "1.0", "description": "The Noromaid 20b model is designed for role-playing and general use, featuring a unique touch with the no_robots dataset that enhances human-like behavior.", "format": "gguf", diff --git a/models/openhermes-neural-7b/model.json b/models/openhermes-neural-7b/model.json index e9e26706c..a7366f85b 100644 --- a/models/openhermes-neural-7b/model.json +++ b/models/openhermes-neural-7b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/OpenHermes-2.5-neural-chat-7B-v3-2-7B-GGUF/resolve/main/openhermes-2.5-neural-chat-7b-v3-2-7b.Q4_K_M.gguf", "id": "openhermes-neural-7b", "object": "model", - "name": "OpenHermes Neural 7B", + "name": "OpenHermes Neural 7B Q4", "version": "1.0", "description": "OpenHermes Neural is a merged model from OpenHermes-2.5-Mistral-7B and neural-chat-7b-v3-2 with the TIES method.", "format": "gguf", diff --git a/models/openorca-13b/model.json b/models/openorca-13b/model.json index 13b6abf29..edfef0807 100644 --- a/models/openorca-13b/model.json +++ b/models/openorca-13b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/Orca-2-13B-GGUF/resolve/main/orca-2-13b.Q5_K_M.gguf", "id": "openorca-13b", "object": "model", - "name": "Orca 2 13B", + "name": "Orca 2 13B Q5", "version": "1.0", "description": "Orca 2 is a finetuned version of LLAMA-2, designed primarily for single-turn responses in reasoning, reading comprehension, math problem solving, and text summarization.", "format": "gguf", diff --git a/models/phind-34b/model.json b/models/phind-34b/model.json index 761c51346..8de7a84d0 100644 --- a/models/phind-34b/model.json +++ b/models/phind-34b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/Phind-CodeLlama-34B-v2-GGUF/resolve/main/phind-codellama-34b-v2.Q5_K_M.gguf", "id": "phind-34b", "object": "model", - "name": "Phind 34B", + "name": "Phind 34B Q5", "version": "1.0", "description": "Phind-CodeLlama-34B-v2 is an AI model fine-tuned on 1.5B tokens of high-quality programming data. It's a SOTA open-source model in coding. This multi-lingual model excels in various programming languages, including Python, C/C++, TypeScript, Java, and is designed to be steerable and user-friendly.", "format": "gguf", diff --git a/models/rocket-3b/model.json b/models/rocket-3b/model.json index a70db1dbd..7435bfafb 100644 --- a/models/rocket-3b/model.json +++ b/models/rocket-3b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/rocket-3B-GGUF/resolve/main/rocket-3b.Q4_K_M.gguf", "id": "rocket-3b", "object": "model", - "name": "Rocket 3B", + "name": "Rocket 3B Q4", "version": "1.0", "description": "Rocket-3B is a GPT-like model, primarily English, fine-tuned on diverse public datasets. It outperforms larger models in benchmarks, showcasing superior understanding and text generation, making it an effective chat model for its size.", "format": "gguf", diff --git a/models/starling-7b/model.json b/models/starling-7b/model.json index cf8aeda18..a7ba963fe 100644 --- a/models/starling-7b/model.json +++ b/models/starling-7b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/Starling-LM-7B-alpha-GGUF/resolve/main/starling-lm-7b-alpha.Q4_K_M.gguf", "id": "starling-7b", "object": "model", - "name": "Strarling alpha 7B", + "name": "Strarling alpha 7B Q4", "version": "1.0", "description": "Starling-RM-7B-alpha is a language model finetuned with Reinforcement Learning from AI Feedback from Openchat 3.5. It stands out for its impressive performance using GPT-4 as a judge, making it one of the top-performing models in its category.", "format": "gguf", diff --git a/models/tiefighter-13b/model.json b/models/tiefighter-13b/model.json index 4dd9bb37f..9940c9a70 100644 --- a/models/tiefighter-13b/model.json +++ b/models/tiefighter-13b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/LLaMA2-13B-Tiefighter-GGUF/resolve/main/llama2-13b-tiefighter.Q5_K_M.gguf", "id": "tiefighter-13b", "object": "model", - "name": "Tiefighter 13B", + "name": "Tiefighter 13B Q5", "version": "1.0", "description": "Tiefighter-13B is a highly creative, merged AI model achieved by combining various 'LORAs' on top of an existing merge, particularly focusing on storytelling and improvisation. This model excels in story writing, chatbots, and adventuring, and is designed to perform better with less detailed inputs, leveraging its inherent creativity.", "format": "gguf", diff --git a/models/tinyllama-1.1b/model.json b/models/tinyllama-1.1b/model.json index 6f10fde02..92793d7f6 100644 --- a/models/tinyllama-1.1b/model.json +++ b/models/tinyllama-1.1b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v0.6/resolve/main/ggml-model-q4_0.gguf", "id": "tinyllama-1.1b", "object": "model", - "name": "TinyLlama Chat 1.1B", + "name": "TinyLlama Chat 1.1B Q4", "version": "1.0", "description": "The TinyLlama project, featuring a 1.1B parameter Llama model, is pretrained on an expansive 3 trillion token dataset. Its design ensures easy integration with various Llama-based open-source projects. Despite its smaller size, it efficiently utilizes lower computational and memory resources, drawing on GPT-4's analytical prowess to enhance its conversational abilities and versatility.", "format": "gguf", diff --git a/models/wizardcoder-13b/model.json b/models/wizardcoder-13b/model.json index 47899534a..a7284aa7c 100644 --- a/models/wizardcoder-13b/model.json +++ b/models/wizardcoder-13b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/WizardCoder-Python-13B-V1.0-GGUF/resolve/main/wizardcoder-python-13b-v1.0.Q5_K_M.gguf", "id": "wizardcoder-13b", "object": "model", - "name": "Wizard Coder Python 13B", + "name": "Wizard Coder Python 13B Q5", "version": "1.0", "description": "WizardCoder-Python-13B is a Python coding model major models like ChatGPT-3.5. This model based on the Llama2 architecture, demonstrate high proficiency in specific domains like coding and mathematics.", "format": "gguf", diff --git a/models/wizardcoder-34b/model.json b/models/wizardcoder-34b/model.json index f564afc35..ca4a3da8e 100644 --- a/models/wizardcoder-34b/model.json +++ b/models/wizardcoder-34b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/WizardCoder-Python-34B-V1.0-GGUF/resolve/main/wizardcoder-python-34b-v1.0.Q5_K_M.gguf", "id": "wizardcoder-34b", "object": "model", - "name": "Wizard Coder Python 34B", + "name": "Wizard Coder Python 34B Q5", "version": "1.0", "description": "WizardCoder-Python-34B is a Python coding model major models like ChatGPT-3.5. This model based on the Llama2 architecture, demonstrate high proficiency in specific domains like coding and mathematics.", "format": "gguf", diff --git a/models/yi-34b/model.json b/models/yi-34b/model.json index d34c0efed..bc6e8e9e3 100644 --- a/models/yi-34b/model.json +++ b/models/yi-34b/model.json @@ -2,7 +2,7 @@ "source_url": "https://huggingface.co/TheBloke/Yi-34B-Chat-GGUF/resolve/main/yi-34b-chat.Q5_K_M.gguf", "id": "yi-34b", "object": "model", - "name": "Yi 34B", + "name": "Yi 34B Q5", "version": "1.0", "description": "Yi-34B, a specialized chat model, is known for its diverse and creative responses and excels across various NLP tasks and benchmarks.", "format": "gguf", diff --git a/models/zephyr-beta-7b/model.json b/models/zephyr-beta-7b/model.json index 70c7c72e8..e1a175432 100644 --- a/models/zephyr-beta-7b/model.json +++ b/models/zephyr-beta-7b/model.json @@ -2,9 +2,9 @@ "source_url": "https://huggingface.co/TheBloke/zephyr-7B-beta-GGUF/resolve/main/zephyr-7b-beta.Q4_K_M.gguf", "id": "zephyr-beta-7b", "object": "model", - "name": "Zephyr Beta 7B", + "name": "Zephyr Beta 7B Q4", "version": "1.0", - "description": "The Zephyr-7B-β model marks the second iteration in the Zephyr series, designed to function as an effective assistant. It has been fine-tuned from the mistralai/Mistral-7B-v0.1 base model, utilizing a combination of public and synthetic datasets with the application of Direct Preference Optimization.", + "description": "The Zephyr-7B-β model is trained by HuggingFace, designed to function as a practical assistant. It has been fine-tuned from the mistralai/Mistral-7B-v0.1 base model, utilizing a combination of public and synthetic datasets with the application of Direct Preference Optimization.", "format": "gguf", "settings": { "ctx_len": 2048,