From fc6f8d2c0a2677660d572e20e067830a905930fe Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 11:41:30 +0700 Subject: [PATCH 01/17] decapriate some 70b models --- models/storytelling-70b/model.json | 24 ------------------------ models/xwin-70b/model.json | 24 ------------------------ models/yarn-70b/model.json | 21 --------------------- 3 files changed, 69 deletions(-) delete mode 100644 models/storytelling-70b/model.json delete mode 100644 models/xwin-70b/model.json delete mode 100644 models/yarn-70b/model.json diff --git a/models/storytelling-70b/model.json b/models/storytelling-70b/model.json deleted file mode 100644 index 76e6f7922..000000000 --- a/models/storytelling-70b/model.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "source_url": "https://huggingface.co/TheBloke/GOAT-70B-Storytelling-GGUF/resolve/main/goat-70b-storytelling.Q5_K_M.gguf", - "id": "storytelling-70b", - "object": "model", - "name": "Storytelling 70B", - "version": "1.0", - "description": "The GOAT-70B-Storytelling model is designed for autonomous story-writing, including crafting books and movie scripts. Based on the LLaMA 2 70B architecture, this model excels in generating cohesive and engaging narratives using inputs like plot outlines and character profiles.", - "format": "gguf", - "settings": { - "ctx_len": 4096, - "system_prompt": "", - "user_prompt": "### USER: ", - "ai_prompt": "\n### ASSISTANT: " - }, - "parameters": { - "max_tokens": 4096 - }, - "metadata": { - "author": "GOAT-AI, The Bloke", - "tags": ["General Use", "Writing"], - "size": 48750000000 - } - } - \ No newline at end of file diff --git a/models/xwin-70b/model.json b/models/xwin-70b/model.json deleted file mode 100644 index a5c1647b0..000000000 --- a/models/xwin-70b/model.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "source_url": "https://huggingface.co/TheBloke/Xwin-LM-70B-V0.1-GGUF/resolve/main/xwin-lm-70b-v0.1.Q5_K_M.gguf", - "id": "xwin-70b", - "object": "model", - "name": "Xwin LM 70B", - "version": "1.0", - "description": "Xwin-LM, based on Llama2 models, emphasizes alignment and exhibits advanced language understanding, text generation, and role-playing abilities.", - "format": "gguf", - "settings": { - "ctx_len": 4096, - "system_prompt": "", - "user_prompt": "USER: ", - "ai_prompt": "ASSISTANT: " - }, - "parameters": { - "max_tokens": 4096 - }, - "metadata": { - "author": "Xwin-LM, The Bloke", - "tags": ["General Use", "Role-playing"], - "size": 48750000000 - } - } - \ No newline at end of file diff --git a/models/yarn-70b/model.json b/models/yarn-70b/model.json deleted file mode 100644 index 67d8d3804..000000000 --- a/models/yarn-70b/model.json +++ /dev/null @@ -1,21 +0,0 @@ -{ - "source_url": "https://huggingface.co/TheBloke/Yarn-Llama-2-70B-32k-GGUF/resolve/main/yarn-llama-2-70b-32k.Q5_K_M.gguf", - "id": "yarn-70b", - "object": "model", - "name": "Yarn 32k 70B", - "version": "1,0", - "description": "Yarn-Llama-2-70b-32k is designed specifically for handling long contexts. It represents an extension of the Llama-2-70b-hf model, now supporting a 32k token context window.", - "format": "gguf", - "settings": { - "ctx_len": 4096 - }, - "parameters": { - "max_tokens": 4096 - }, - "metadata": { - "author": "NousResearch, The Bloke", - "tags": ["General Use", "Big Context Length"], - "size": 48750000000 - } - } - \ No newline at end of file From 2e96aacf55b4061dfb5ca5560e9691484a659728 Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 11:42:46 +0700 Subject: [PATCH 02/17] decapriate some 34b models --- models/dolphin-yi-34b/model.json | 24 ------------------------ models/lzlv-70b/model.json | 2 +- 2 files changed, 1 insertion(+), 25 deletions(-) delete mode 100644 models/dolphin-yi-34b/model.json diff --git a/models/dolphin-yi-34b/model.json b/models/dolphin-yi-34b/model.json deleted file mode 100644 index 3b1bf3619..000000000 --- a/models/dolphin-yi-34b/model.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "source_url": "https://huggingface.co/TheBloke/dolphin-2_2-yi-34b-GGUF/resolve/main/dolphin-2_2-yi-34b.Q5_K_M.gguf", - "id": "dolphin-yi-34b", - "object": "model", - "name": "Dolphin Yi 34B", - "version": "1.0", - "description": "Dolphin, based on the Yi-34B model and enhanced with features like conversation and empathy, is trained on a unique dataset for advanced multi-turn conversations. Notably uncensored, it requires careful implementation of an alignment layer for ethical use.", - "format": "gguf", - "settings": { - "ctx_len": 4096, - "system_prompt": "<|im_start|>system\n", - "user_prompt": "<|im_end|>\n<|im_start|>user\n", - "ai_prompt": "<|im_end|>\n<|im_start|>assistant\n" - }, - "parameters": { - "max_tokens": 4096 - }, - "metadata": { - "author": "ehartford, The Bloke", - "tags": ["General Use", "Role-playing"], - "size": 24320000000 - } - } - \ No newline at end of file diff --git a/models/lzlv-70b/model.json b/models/lzlv-70b/model.json index 4aaee79b3..166e3d8ed 100644 --- a/models/lzlv-70b/model.json +++ b/models/lzlv-70b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "lizpreciatior, The Bloke", - "tags": ["General Use", "Role-playing"], + "tags": ["Community Recommended", "General Use", "Role-playing"], "size": 48750000000 } } From 5cc8e0f9a9c5cc16ea2769a5337b9cab56f2ea87 Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 11:50:56 +0700 Subject: [PATCH 03/17] decapriate some small models --- models/capybara-34b/model.json | 4 ++-- models/deepseek-coder-1.3b/model.json | 2 +- models/islm-3b/model.json | 24 ------------------------ models/marx-3b/model.json | 23 ----------------------- models/mythomax-13b/model.json | 24 ------------------------ 5 files changed, 3 insertions(+), 74 deletions(-) delete mode 100644 models/islm-3b/model.json delete mode 100644 models/marx-3b/model.json delete mode 100644 models/mythomax-13b/model.json diff --git a/models/capybara-34b/model.json b/models/capybara-34b/model.json index 562bcbe93..d2da8d002 100644 --- a/models/capybara-34b/model.json +++ b/models/capybara-34b/model.json @@ -3,7 +3,7 @@ "id": "capybara-34b", "object": "model", "name": "Capybara 200k 34B", - "version": 1.0, + "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", "settings": { @@ -17,7 +17,7 @@ }, "metadata": { "author": "NousResearch, The Bloke", - "tags": ["General", "Big Context Length"], + "tags": ["Community Recommended", "General Use", "Big Context Length"], "size": 24320000000 } } diff --git a/models/deepseek-coder-1.3b/model.json b/models/deepseek-coder-1.3b/model.json index 2ff6d6e7b..3ee705d20 100644 --- a/models/deepseek-coder-1.3b/model.json +++ b/models/deepseek-coder-1.3b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "deepseek, The Bloke", - "tags": ["Code"], + "tags": ["Community Recommended", "Code", "Small size"], "size": 870000000 } } diff --git a/models/islm-3b/model.json b/models/islm-3b/model.json deleted file mode 100644 index 916d7c50e..000000000 --- a/models/islm-3b/model.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "source_url": "https://huggingface.co/UmbrellaCorp/IS-LM-3B_GGUF/resolve/main/IS-LM-Q4_K_M.gguf", - "id": "islm-3b", - "object": "model", - "name": "IS LM 3B", - "version": "1.0", - "description": "IS LM 3B, based on the StableLM 3B model is specifically finetuned for economic analysis using DataForge Economics and QLoRA over three epochs, enhancing its proficiency in economic forecasting and analysis.", - "format": "gguf", - "settings": { - "ctx_len": 4096, - "system_prompt": "", - "user_prompt": "USER: ", - "ai_prompt": "ASSISTANT: " - }, - "parameters": { - "max_tokens": 4096 - }, - "metadata": { - "author": "UmbrellaCorp, The Bloke", - "tags": ["General Use", "Economics"], - "size": 1710000000 - } - } - \ No newline at end of file diff --git a/models/marx-3b/model.json b/models/marx-3b/model.json deleted file mode 100644 index 78617d5c3..000000000 --- a/models/marx-3b/model.json +++ /dev/null @@ -1,23 +0,0 @@ -{ - "source_url": "https://huggingface.co/TheBloke/Marx-3B-v3-GGUF/resolve/main/marx-3b-v3.Q4_K_M.gguf", - "id": "marx-3b", - "object": "model", - "name": "Marx 3B", - "version": "1.0", - "description": "Marx 3B, based on the StableLM 3B model is specifically finetuned for chating using EverythingLM data and QLoRA over two epochs, enhancing its proficiency in understand general knowledege.", - "format": "gguf", - "settings": { - "ctx_len": 4096, - "system_prompt": "", - "user_prompt": "### HUMAN: ", - "ai_prompt": "### RESPONSE: " - }, - "parameters": { - "max_tokens": 4096 - }, - "metadata": { - "author": "Bohan Du, The Bloke", - "tags": ["General Use"], - "size": 1620000000 - } - } \ No newline at end of file diff --git a/models/mythomax-13b/model.json b/models/mythomax-13b/model.json deleted file mode 100644 index 455f73968..000000000 --- a/models/mythomax-13b/model.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "source_url": "https://huggingface.co/TheBloke/MythoMax-L2-13B-GGUF/resolve/main/mythomax-l2-13b.Q5_K_M.gguf", - "id": "mythomax-13b", - "object": "model", - "name": "Mythomax L2 13B", - "version": "1.0", - "description": "Mythomax L2 13b, an advanced AI model derived from MythoMix, merges MythoLogic-L2's deep comprehension with Huginn's writing skills through a unique tensor merge technique, excelling in roleplaying and storytelling.", - "format": "gguf", - "settings": { - "ctx_len": 4096, - "system_prompt": "", - "user_prompt": "### Instruction: ", - "ai_prompt": "### Response: " - }, - "parameters": { - "max_tokens": 4096 - }, - "metadata": { - "author": "Gryphe, The Bloke", - "tags": ["Role-playing"], - "size": 9230000000 - } - } - \ No newline at end of file From 4440b58b15668b6538b87fa75b500e0d5c299815 Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 11:52:34 +0700 Subject: [PATCH 04/17] decapriate some 7B models --- models/openchat-7b/model.json | 24 ------------------------ models/openhermes-mistral-7b/model.json | 24 ------------------------ models/openorca-7b/model.json | 24 ------------------------ 3 files changed, 72 deletions(-) delete mode 100644 models/openchat-7b/model.json delete mode 100644 models/openhermes-mistral-7b/model.json delete mode 100644 models/openorca-7b/model.json diff --git a/models/openchat-7b/model.json b/models/openchat-7b/model.json deleted file mode 100644 index 1fd6bb259..000000000 --- a/models/openchat-7b/model.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "source_url": "https://huggingface.co/TheBloke/openchat_3.5-GGUF/resolve/main/openchat_3.5.Q4_K_M.gguf", - "id": "openchat-7b", - "object": "model", - "name": "Open Chat 3.5 7B", - "version": "1.0", - "description": "OpenChat represents a breakthrough in the realm of open-source language models. By implementing the C-RLFT fine-tuning strategy, inspired by offline reinforcement learning, this 7B model achieves results on par with ChatGPT (March).", - "format": "gguf", - "settings": { - "ctx_len": 4096, - "system_prompt": "", - "user_prompt": "GPT4 User: ", - "ai_prompt": "<|end_of_turn|>\nGPT4 Assistant: " - }, - "parameters": { - "max_tokens": 4096 - }, - "metadata": { - "author": "OpenChat, The Bloke", - "tags": ["General", "Code"], - "size": 4370000000 - } - } - \ No newline at end of file diff --git a/models/openhermes-mistral-7b/model.json b/models/openhermes-mistral-7b/model.json deleted file mode 100644 index 6b64363d5..000000000 --- a/models/openhermes-mistral-7b/model.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "source_url": "https://huggingface.co/TheBloke/OpenHermes-2.5-Mistral-7B-GGUF/resolve/main/openhermes-2.5-mistral-7b.Q4_K_M.gguf", - "id": "openhermes-mistral-7b", - "object": "model", - "name": "Openhermes 2.5 Mistral 7B", - "version": "1.0", - "description": "The OpenHermes 2.5 Mistral 7B incorporates additional code datasets, more than a million GPT-4 generated data examples, and other high-quality open datasets. This enhancement led to significant improvement in benchmarks, highlighting its improved skill in handling code-centric tasks.", - "format": "gguf", - "settings": { - "ctx_len": 4096, - "system_prompt": "<|im_start|>system\n", - "user_prompt": "<|im_end|>\n<|im_start|>user\n", - "ai_prompt": "<|im_end|>\n<|im_start|>assistant\n" - }, - "parameters": { - "max_tokens": 4096 - }, - "metadata": { - "author": "Teknium, The Bloke", - "tags": ["General", "Roleplay"], - "size": 4370000000 - } - } - \ No newline at end of file diff --git a/models/openorca-7b/model.json b/models/openorca-7b/model.json deleted file mode 100644 index 42c88212c..000000000 --- a/models/openorca-7b/model.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "source_url": "https://huggingface.co/TheBloke/Mistral-7B-OpenOrca-GGUF/resolve/main/mistral-7b-openorca.Q4_K_M.gguf", - "id": "openorca-7b", - "object": "model", - "name": "OpenOrca 7B", - "version": "1.0", - "description": "OpenOrca 8k 7B is a model based on Mistral 7B, fine-tuned using the OpenOrca dataset. Notably ranked first on the HF Leaderboard for models under 30B, it excels in efficiency and accessibility.", - "format": "gguf", - "settings": { - "ctx_len": 4096, - "system_prompt": "<|im_start|>system\n", - "user_prompt": "<|im_end|>\n<|im_start|>user\n", - "ai_prompt": "<|im_end|>\n<|im_start|>assistant\n" - }, - "parameters": { - "max_tokens": 4096 - }, - "metadata": { - "author": "OpenOrca, The Bloke", - "tags": ["General", "Code"], - "size": 4370000000 - } - } - \ No newline at end of file From c27a1cc3796e86b2afe536a48c75c53459e6d5c3 Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 11:56:39 +0700 Subject: [PATCH 05/17] update tags for models --- models/neural-chat-7b/model.json | 2 +- models/neuralhermes-7b/model.json | 2 +- models/noromaid-20b/model.json | 2 +- models/openorca-13b/model.json | 2 +- models/phind-34b/model.json | 2 +- models/rocket-3b/model.json | 2 +- models/starling-7b/model.json | 2 +- models/tiefighter-13b/model.json | 2 +- models/tinyllama-1.1b/model.json | 2 +- models/wizardcoder-13b/model.json | 2 +- models/wizardcoder-34b/model.json | 2 +- models/yi-34b/model.json | 2 +- models/zephyr-beta-7b/model.json | 2 +- 13 files changed, 13 insertions(+), 13 deletions(-) diff --git a/models/neural-chat-7b/model.json b/models/neural-chat-7b/model.json index f4f4f14d4..a31662496 100644 --- a/models/neural-chat-7b/model.json +++ b/models/neural-chat-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Intel, The Bloke", - "tags": ["General Use", "Role-playing", "Big Context Length"], + "tags": ["Community Recommended", "General Use", "Role-playing", "Big Context Length"], "size": 4370000000 } } diff --git a/models/neuralhermes-7b/model.json b/models/neuralhermes-7b/model.json index 07cca58d4..3ec0c6253 100644 --- a/models/neuralhermes-7b/model.json +++ b/models/neuralhermes-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Intel, The Bloke", - "tags": ["General Use", "Code", "Big Context Length"], + "tags": ["Community Recommended", "General Use", "Code", "Big Context Length"], "size": 4370000000 } } diff --git a/models/noromaid-20b/model.json b/models/noromaid-20b/model.json index 86291e4f5..f0460568c 100644 --- a/models/noromaid-20b/model.json +++ b/models/noromaid-20b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "NeverSleep, The Bloke", - "tags": ["Role-playing"], + "tags": ["Community Recommended", "Role-playing"], "size": 12040000000 } } diff --git a/models/openorca-13b/model.json b/models/openorca-13b/model.json index 02a555430..5027dbd22 100644 --- a/models/openorca-13b/model.json +++ b/models/openorca-13b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Microsoft, The Bloke", - "tags": ["General Use"], + "tags": ["Community Recommended", "General Use"], "size": 9230000000 } } diff --git a/models/phind-34b/model.json b/models/phind-34b/model.json index 4391ae08d..18ed50bcc 100644 --- a/models/phind-34b/model.json +++ b/models/phind-34b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Phind, The Bloke", - "tags": ["Code", "Big Context Length"], + "tags": ["Community Recommended", "Code", "Big Context Length"], "size": 24320000000 } } diff --git a/models/rocket-3b/model.json b/models/rocket-3b/model.json index b00eb1f44..4eddb04b0 100644 --- a/models/rocket-3b/model.json +++ b/models/rocket-3b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "pansophic, The Bloke", - "tags": ["General Use"], + "tags": ["Community Recommended", "General Use"], "size": 1710000000 } } \ No newline at end of file diff --git a/models/starling-7b/model.json b/models/starling-7b/model.json index c029ea7d9..4ff5651e0 100644 --- a/models/starling-7b/model.json +++ b/models/starling-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Berkeley-nest, The Bloke", - "tags": ["General", "Code"], + "tags": ["Community Recommended", "General", "Code"], "size": 4370000000 } } diff --git a/models/tiefighter-13b/model.json b/models/tiefighter-13b/model.json index b1d354ce3..ea38fadbf 100644 --- a/models/tiefighter-13b/model.json +++ b/models/tiefighter-13b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "KoboldAI, The Bloke", - "tags": ["General Use", "Role-playing", "Writing"], + "tags": ["Community Recommended", "General Use", "Role-playing", "Writing"], "size": 9230000000 } } diff --git a/models/tinyllama-1.1b/model.json b/models/tinyllama-1.1b/model.json index f561eb25d..6f10fde02 100644 --- a/models/tinyllama-1.1b/model.json +++ b/models/tinyllama-1.1b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "TinyLlama", - "tags": ["General Use"], + "tags": ["Community Recommended", "General Use"], "size": 637000000 } } \ No newline at end of file diff --git a/models/wizardcoder-13b/model.json b/models/wizardcoder-13b/model.json index 944b5632b..9fe250348 100644 --- a/models/wizardcoder-13b/model.json +++ b/models/wizardcoder-13b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "WizardLM, The Bloke", - "tags": ["Code", "Big Context Length"], + "tags": ["Community Recommended", "Code", "Big Context Length"], "size": 9230000000 } } diff --git a/models/wizardcoder-34b/model.json b/models/wizardcoder-34b/model.json index aa2618e1b..db92a4fbf 100644 --- a/models/wizardcoder-34b/model.json +++ b/models/wizardcoder-34b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "WizardLM, The Bloke", - "tags": ["Code", "Big Context Length"], + "tags": ["Community Recommended", "Code", "Big Context Length"], "size": 24320000000 } } diff --git a/models/yi-34b/model.json b/models/yi-34b/model.json index f899bc54b..9bbe65760 100644 --- a/models/yi-34b/model.json +++ b/models/yi-34b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "01-ai, The Bloke", - "tags": ["General", "Role-playing", "Writing"], + "tags": ["Foundational Model", "General Use", "Role-playing", "Writing"], "size": 24320000000 } } diff --git a/models/zephyr-beta-7b/model.json b/models/zephyr-beta-7b/model.json index 24529bc9a..474dba766 100644 --- a/models/zephyr-beta-7b/model.json +++ b/models/zephyr-beta-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "HuggingFaceH4, The Bloke", - "tags": ["General Use", "Big Context Length"], + "tags": ["Community Recommended", "General Use", "Big Context Length"], "size": 4370000000 } } From 6a5458731991866cc2dafffcbbab9477d2283cf0 Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 12:26:50 +0700 Subject: [PATCH 06/17] update llama 2 7b chat --- models/capybara-34b/model.json | 4 ++-- models/deepseek-coder-1.3b/model.json | 4 ++-- models/llama2-chat-7b-q4/model.json | 24 ++++++++++++++++++++++++ models/llama2-chat-7b-q5/model.json | 24 ++++++++++++++++++++++++ 4 files changed, 52 insertions(+), 4 deletions(-) create mode 100644 models/llama2-chat-7b-q4/model.json create mode 100644 models/llama2-chat-7b-q5/model.json diff --git a/models/capybara-34b/model.json b/models/capybara-34b/model.json index d2da8d002..7c669bbb7 100644 --- a/models/capybara-34b/model.json +++ b/models/capybara-34b/model.json @@ -7,13 +7,13 @@ "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", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "", "user_prompt": "USER: ", "ai_prompt": "ASSISTANT: " }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "NousResearch, The Bloke", diff --git a/models/deepseek-coder-1.3b/model.json b/models/deepseek-coder-1.3b/model.json index 3ee705d20..dac90423d 100644 --- a/models/deepseek-coder-1.3b/model.json +++ b/models/deepseek-coder-1.3b/model.json @@ -7,13 +7,13 @@ "description": "", "format": "gguf", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "", "user_prompt": "", "ai_prompt": "" }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "deepseek, The Bloke", diff --git a/models/llama2-chat-7b-q4/model.json b/models/llama2-chat-7b-q4/model.json new file mode 100644 index 000000000..be302d144 --- /dev/null +++ b/models/llama2-chat-7b-q4/model.json @@ -0,0 +1,24 @@ +{ + "source_url": "https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGUF/resolve/main/llama-2-7b-chat.Q4_K_M.gguf", + "id": "llama2-chat-7b-q4", + "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.", + "format": "gguf", + "settings": { + "ctx_len": 2048, + "system_prompt": "[INST] <>\n", + "user_prompt": "<>\n", + "ai_prompt": "[/INST]" + }, + "parameters": { + "max_tokens": 2048 + }, + "metadata": { + "author": "MetaAI, The Bloke", + "tags": ["Foundational Model", "General", "Code"], + "size": 4080000000 + } + } + \ No newline at end of file diff --git a/models/llama2-chat-7b-q5/model.json b/models/llama2-chat-7b-q5/model.json new file mode 100644 index 000000000..8a93327d6 --- /dev/null +++ b/models/llama2-chat-7b-q5/model.json @@ -0,0 +1,24 @@ +{ + "source_url": "https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGUF/resolve/main/llama-2-7b-chat.Q5_K_M.gguf", + "id": "llama2-chat-7b-q5", + "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.", + "format": "gguf", + "settings": { + "ctx_len": 2048, + "system_prompt": "[INST] <>\n", + "user_prompt": "<>\n", + "ai_prompt": "[/INST]" + }, + "parameters": { + "max_tokens": 2048 + }, + "metadata": { + "author": "MetaAI, The Bloke", + "tags": ["Foundational Model", "General", "Code"], + "size": 4780000000 + } + } + \ No newline at end of file From dbc12d27c89f31e1939150344cac57885c8025ee Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 12:32:56 +0700 Subject: [PATCH 07/17] update mistral 7b chat --- models/lzlv-70b/model.json | 4 ++-- models/mistral-ins-7b-q4/model.json | 24 ++++++++++++++++++++++++ models/mistral-ins-7b-q5/model.json | 24 ++++++++++++++++++++++++ 3 files changed, 50 insertions(+), 2 deletions(-) create mode 100644 models/mistral-ins-7b-q4/model.json create mode 100644 models/mistral-ins-7b-q5/model.json diff --git a/models/lzlv-70b/model.json b/models/lzlv-70b/model.json index 166e3d8ed..98d581f24 100644 --- a/models/lzlv-70b/model.json +++ b/models/lzlv-70b/model.json @@ -7,13 +7,13 @@ "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", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "", "user_prompt": "USER: ", "ai_prompt": "ASSISTANT: " }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "lizpreciatior, The Bloke", diff --git a/models/mistral-ins-7b-q4/model.json b/models/mistral-ins-7b-q4/model.json new file mode 100644 index 000000000..51b5c231e --- /dev/null +++ b/models/mistral-ins-7b-q4/model.json @@ -0,0 +1,24 @@ +{ + "source_url": "https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF/resolve/main/mistral-7b-instruct-v0.1.Q4_K_M.gguf", + "id": "mistral-ins-7b-q4", + "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.", + "format": "gguf", + "settings": { + "ctx_len": 2048, + "system_prompt": "", + "user_prompt": "[INST]", + "ai_prompt": "[/INST]" + }, + "parameters": { + "max_tokens": 2048 + }, + "metadata": { + "author": "MistralAI, The Bloke", + "tags": ["Foundational Model", "General", "Code"], + "size": 4370000000 + } + } + \ No newline at end of file diff --git a/models/mistral-ins-7b-q5/model.json b/models/mistral-ins-7b-q5/model.json new file mode 100644 index 000000000..46c5f5378 --- /dev/null +++ b/models/mistral-ins-7b-q5/model.json @@ -0,0 +1,24 @@ +{ + "source_url": "https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF/resolve/main/mistral-7b-instruct-v0.1.Q5_K_M.gguf", + "id": "mistral-ins-7b-q5", + "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.", + "format": "gguf", + "settings": { + "ctx_len": 2048, + "system_prompt": "", + "user_prompt": "[INST]", + "ai_prompt": "[/INST]" + }, + "parameters": { + "max_tokens": 2048 + }, + "metadata": { + "author": "MistralAI, The Bloke", + "tags": ["Foundational Model", "General", "Code"], + "size": 5130000000 + } + } + \ No newline at end of file From 88e37e1f2d6d3b0c0e0f91bcf615233a2ccdbd2f Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 12:34:10 +0700 Subject: [PATCH 08/17] update ctxlen --- models/neural-chat-7b/model.json | 4 ++-- models/neuralhermes-7b/model.json | 4 ++-- models/noromaid-20b/model.json | 4 ++-- models/openorca-13b/model.json | 4 ++-- models/phind-34b/model.json | 4 ++-- models/rocket-3b/model.json | 2 +- models/starling-7b/model.json | 4 ++-- models/tiefighter-13b/model.json | 4 ++-- models/wizardcoder-13b/model.json | 4 ++-- models/wizardcoder-34b/model.json | 4 ++-- models/yi-34b/model.json | 4 ++-- models/zephyr-beta-7b/model.json | 4 ++-- 12 files changed, 23 insertions(+), 23 deletions(-) diff --git a/models/neural-chat-7b/model.json b/models/neural-chat-7b/model.json index a31662496..84c8d029f 100644 --- a/models/neural-chat-7b/model.json +++ b/models/neural-chat-7b/model.json @@ -7,13 +7,13 @@ "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", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "### System: ", "user_prompt": "### User: ", "ai_prompt": "### Assistant: " }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "Intel, The Bloke", diff --git a/models/neuralhermes-7b/model.json b/models/neuralhermes-7b/model.json index 3ec0c6253..0cc83d7fb 100644 --- a/models/neuralhermes-7b/model.json +++ b/models/neuralhermes-7b/model.json @@ -7,13 +7,13 @@ "description": "NeuralHermes 2.5 has been enhanced using Direct Preference Optimization. This fine-tuning, inspired by the RLHF process of Neural-chat-7b and OpenHermes-2.5-Mistral-7B, has led to improved performance across several benchmarks.", "format": "gguf", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "<|im_start|>system\n", "user_prompt": "<|im_end|>\n<|im_start|>user\n", "ai_prompt": "<|im_end|>\n<|im_start|>assistant\n" }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "Intel, The Bloke", diff --git a/models/noromaid-20b/model.json b/models/noromaid-20b/model.json index f0460568c..ff2b91c1d 100644 --- a/models/noromaid-20b/model.json +++ b/models/noromaid-20b/model.json @@ -7,13 +7,13 @@ "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", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "", "user_prompt": "### Instruction: ", "ai_prompt": "### Response: " }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "NeverSleep, The Bloke", diff --git a/models/openorca-13b/model.json b/models/openorca-13b/model.json index 5027dbd22..13b6abf29 100644 --- a/models/openorca-13b/model.json +++ b/models/openorca-13b/model.json @@ -7,13 +7,13 @@ "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", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "<|im_start|>system\n", "user_prompt": "<|im_end|>\n<|im_start|>user\n", "ai_prompt": "<|im_end|>\n<|im_start|>assistant\n" }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "Microsoft, The Bloke", diff --git a/models/phind-34b/model.json b/models/phind-34b/model.json index 18ed50bcc..761c51346 100644 --- a/models/phind-34b/model.json +++ b/models/phind-34b/model.json @@ -7,13 +7,13 @@ "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", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "### System Prompt\n", "user_prompt": "### User Message\n", "ai_prompt": "### Assistant\n" }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "Phind, The Bloke", diff --git a/models/rocket-3b/model.json b/models/rocket-3b/model.json index 4eddb04b0..a70db1dbd 100644 --- a/models/rocket-3b/model.json +++ b/models/rocket-3b/model.json @@ -13,7 +13,7 @@ "ai_prompt": "<|im_end|>\n<|im_start|>assistant\n" }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "pansophic, The Bloke", diff --git a/models/starling-7b/model.json b/models/starling-7b/model.json index 4ff5651e0..cf8aeda18 100644 --- a/models/starling-7b/model.json +++ b/models/starling-7b/model.json @@ -7,13 +7,13 @@ "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", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "", "user_prompt": "GPT4 User: ", "ai_prompt": "<|end_of_turn|>\nGPT4 Assistant: " }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "Berkeley-nest, The Bloke", diff --git a/models/tiefighter-13b/model.json b/models/tiefighter-13b/model.json index ea38fadbf..4dd9bb37f 100644 --- a/models/tiefighter-13b/model.json +++ b/models/tiefighter-13b/model.json @@ -7,13 +7,13 @@ "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", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "", "user_prompt": "### Instruction: ", "ai_prompt": "\n### Response: " }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "KoboldAI, The Bloke", diff --git a/models/wizardcoder-13b/model.json b/models/wizardcoder-13b/model.json index 9fe250348..47899534a 100644 --- a/models/wizardcoder-13b/model.json +++ b/models/wizardcoder-13b/model.json @@ -7,13 +7,13 @@ "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", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "", "user_prompt": "### Instruction:\n", "ai_prompt": "### Response:\n" }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "WizardLM, The Bloke", diff --git a/models/wizardcoder-34b/model.json b/models/wizardcoder-34b/model.json index db92a4fbf..f564afc35 100644 --- a/models/wizardcoder-34b/model.json +++ b/models/wizardcoder-34b/model.json @@ -7,13 +7,13 @@ "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", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "", "user_prompt": "### Instruction:\n", "ai_prompt": "### Response:\n" }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "WizardLM, The Bloke", diff --git a/models/yi-34b/model.json b/models/yi-34b/model.json index 9bbe65760..d34c0efed 100644 --- a/models/yi-34b/model.json +++ b/models/yi-34b/model.json @@ -7,13 +7,13 @@ "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", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "<|im_start|>system\n", "user_prompt": "<|im_end|>\n<|im_start|>user\n", "ai_prompt": "<|im_end|>\n<|im_start|>assistant\n" }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "01-ai, The Bloke", diff --git a/models/zephyr-beta-7b/model.json b/models/zephyr-beta-7b/model.json index 474dba766..70c7c72e8 100644 --- a/models/zephyr-beta-7b/model.json +++ b/models/zephyr-beta-7b/model.json @@ -7,13 +7,13 @@ "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.", "format": "gguf", "settings": { - "ctx_len": 4096, + "ctx_len": 2048, "system_prompt": "<|system|>\n", "user_prompt": "\n<|user|>\n", "ai_prompt": "\n<|assistant|>\n" }, "parameters": { - "max_tokens": 4096 + "max_tokens": 2048 }, "metadata": { "author": "HuggingFaceH4, The Bloke", From fc2af712e405d977d33f2873ec6f18ddb1062938 Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 12:35:53 +0700 Subject: [PATCH 09/17] update llama2 70b chat --- models/llama2-chat-70b-q4/model.json | 24 ++++++++++++++++++++++++ 1 file changed, 24 insertions(+) create mode 100644 models/llama2-chat-70b-q4/model.json diff --git a/models/llama2-chat-70b-q4/model.json b/models/llama2-chat-70b-q4/model.json new file mode 100644 index 000000000..39c0f6474 --- /dev/null +++ b/models/llama2-chat-70b-q4/model.json @@ -0,0 +1,24 @@ +{ + "source_url": "https://huggingface.co/TheBloke/Llama-2-70B-Chat-GGUF/resolve/main/llama-2-70b-chat.Q4_K_M.gguf", + "id": "llama2-chat-70b-q4", + "object": "model", + "name": "Llama 2 Chat 70B Q4", + "version": "1.0", + "description": "This is a 4-bit quantized version of Meta AI's Llama 2 Chat 70b model.", + "format": "gguf", + "settings": { + "ctx_len": 2048, + "system_prompt": "[INST] <>\n", + "user_prompt": "<>\n", + "ai_prompt": "[/INST]" + }, + "parameters": { + "max_tokens": 2048 + }, + "metadata": { + "author": "MetaAI, The Bloke", + "tags": ["Foundational Model", "General", "Code"], + "size": 4080000000 + } + } + \ No newline at end of file From 43b538e800a7469a3f25eceab1d97e8a7eb61c90 Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 14:13:00 +0700 Subject: [PATCH 10/17] update openhermes 7b --- models/capybara-34b/model.json | 4 ++-- models/lzlv-70b/model.json | 4 ++-- models/neural-chat-7b/model.json | 6 +++--- .../{neuralhermes-7b => openhermes-neural-7b}/model.json | 8 ++++---- 4 files changed, 11 insertions(+), 11 deletions(-) rename models/{neuralhermes-7b => openhermes-neural-7b}/model.json (52%) diff --git a/models/capybara-34b/model.json b/models/capybara-34b/model.json index 7c669bbb7..74eb27eb9 100644 --- a/models/capybara-34b/model.json +++ b/models/capybara-34b/model.json @@ -9,8 +9,8 @@ "settings": { "ctx_len": 2048, "system_prompt": "", - "user_prompt": "USER: ", - "ai_prompt": "ASSISTANT: " + "user_prompt": "USER:\n", + "ai_prompt": "ASSISTANT:\n" }, "parameters": { "max_tokens": 2048 diff --git a/models/lzlv-70b/model.json b/models/lzlv-70b/model.json index 98d581f24..844bdadf4 100644 --- a/models/lzlv-70b/model.json +++ b/models/lzlv-70b/model.json @@ -9,8 +9,8 @@ "settings": { "ctx_len": 2048, "system_prompt": "", - "user_prompt": "USER: ", - "ai_prompt": "ASSISTANT: " + "user_prompt": "USER:\n", + "ai_prompt": "ASSISTANT:\n" }, "parameters": { "max_tokens": 2048 diff --git a/models/neural-chat-7b/model.json b/models/neural-chat-7b/model.json index 84c8d029f..cac717b8a 100644 --- a/models/neural-chat-7b/model.json +++ b/models/neural-chat-7b/model.json @@ -8,9 +8,9 @@ "format": "gguf", "settings": { "ctx_len": 2048, - "system_prompt": "### System: ", - "user_prompt": "### User: ", - "ai_prompt": "### Assistant: " + "system_prompt": "### System:\n", + "user_prompt": "### User:\n", + "ai_prompt": "### Assistant:\n" }, "parameters": { "max_tokens": 2048 diff --git a/models/neuralhermes-7b/model.json b/models/openhermes-neural-7b/model.json similarity index 52% rename from models/neuralhermes-7b/model.json rename to models/openhermes-neural-7b/model.json index 0cc83d7fb..cd4394dc5 100644 --- a/models/neuralhermes-7b/model.json +++ b/models/openhermes-neural-7b/model.json @@ -1,10 +1,10 @@ { - "source_url": "https://huggingface.co/TheBloke/NeuralHermes-2.5-Mistral-7B-GGUF/resolve/main/neuralhermes-2.5-mistral-7b.Q4_K_M.gguf", - "id": "neuralhermes-7b", + "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": "NeuralHermes 7B", + "name": "OpenHermes Neural 7B", "version": "1.0", - "description": "NeuralHermes 2.5 has been enhanced using Direct Preference Optimization. This fine-tuning, inspired by the RLHF process of Neural-chat-7b and OpenHermes-2.5-Mistral-7B, has led to improved performance across several benchmarks.", + "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", "settings": { "ctx_len": 2048, From 23a6ecdd890be73e04d0ffe990fc81e9b06d8fe0 Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 14:14:43 +0700 Subject: [PATCH 11/17] update tags --- models/openhermes-neural-7b/model.json | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/models/openhermes-neural-7b/model.json b/models/openhermes-neural-7b/model.json index cd4394dc5..e9e26706c 100644 --- a/models/openhermes-neural-7b/model.json +++ b/models/openhermes-neural-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Intel, The Bloke", - "tags": ["Community Recommended", "General Use", "Code", "Big Context Length"], + "tags": ["Community Recommended", "General Use", "Code", "Merged"], "size": 4370000000 } } From 84c940be47d118759a129b66ff80eb2cae1659c2 Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 14:28:36 +0700 Subject: [PATCH 12/17] update prompt templte --- models/noromaid-20b/model.json | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/models/noromaid-20b/model.json b/models/noromaid-20b/model.json index ff2b91c1d..61870e855 100644 --- a/models/noromaid-20b/model.json +++ b/models/noromaid-20b/model.json @@ -9,8 +9,8 @@ "settings": { "ctx_len": 2048, "system_prompt": "", - "user_prompt": "### Instruction: ", - "ai_prompt": "### Response: " + "user_prompt": "### Instruction:\n", + "ai_prompt": "### Response:\n" }, "parameters": { "max_tokens": 2048 From 7017e6426d4ea022a6bc6197c531e3560fc3767a Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 16:22:42 +0700 Subject: [PATCH 13/17] update name for models --- models/capybara-34b/model.json | 2 +- models/deepseek-coder-1.3b/model.json | 4 ++-- models/llama2-chat-70b-q4/model.json | 2 +- models/llama2-chat-7b-q4/model.json | 2 +- models/llama2-chat-7b-q5/model.json | 2 +- models/lzlv-70b/model.json | 2 +- models/mistral-ins-7b-q4/model.json | 2 +- models/mistral-ins-7b-q5/model.json | 2 +- models/neural-chat-7b/model.json | 2 +- models/noromaid-20b/model.json | 2 +- models/openhermes-neural-7b/model.json | 2 +- models/openorca-13b/model.json | 2 +- models/phind-34b/model.json | 2 +- models/rocket-3b/model.json | 2 +- models/starling-7b/model.json | 2 +- models/tiefighter-13b/model.json | 2 +- models/tinyllama-1.1b/model.json | 2 +- models/wizardcoder-13b/model.json | 2 +- models/wizardcoder-34b/model.json | 2 +- models/yi-34b/model.json | 2 +- models/zephyr-beta-7b/model.json | 4 ++-- 21 files changed, 23 insertions(+), 23 deletions(-) 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, From f4dbec7e36527f4f022c55fbc1d426a4039d8851 Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 16:31:47 +0700 Subject: [PATCH 14/17] update deepseek coder --- models/deepseek-coder-1.3b/model.json | 2 +- models/deepseek-coder-34b/model.json | 24 ++++++++++++++++++++++++ models/wizardcoder-34b/model.json | 24 ------------------------ 3 files changed, 25 insertions(+), 25 deletions(-) create mode 100644 models/deepseek-coder-34b/model.json delete mode 100644 models/wizardcoder-34b/model.json diff --git a/models/deepseek-coder-1.3b/model.json b/models/deepseek-coder-1.3b/model.json index 9d22f9558..2bb9e7fee 100644 --- a/models/deepseek-coder-1.3b/model.json +++ b/models/deepseek-coder-1.3b/model.json @@ -4,7 +4,7 @@ "object": "model", "name": "Deepseek Coder 1.3B Q4", "version": "1.0", - "description": "", + "description": "Deepseek Coder trained on 2T tokens (87% code, 13% English/Chinese), excelling in project-level code completion with advanced capabilities across multiple programming languages.", "format": "gguf", "settings": { "ctx_len": 2048, diff --git a/models/deepseek-coder-34b/model.json b/models/deepseek-coder-34b/model.json new file mode 100644 index 000000000..e175396f5 --- /dev/null +++ b/models/deepseek-coder-34b/model.json @@ -0,0 +1,24 @@ +{ + "source_url": "https://huggingface.co/TheBloke/deepseek-coder-33B-instruct-GGUF/resolve/main/deepseek-coder-33b-instruct.Q5_K_M.gguf", + "id": "deepseek-coder-34b", + "object": "model", + "name": "Deepseek Coder 33B Q5", + "version": "1.0", + "description": "Deepseek Coder trained on 2T tokens (87% code, 13% English/Chinese), excelling in project-level code completion with advanced capabilities across multiple programming languages.", + "format": "gguf", + "settings": { + "ctx_len": 2048, + "system_prompt": "", + "user_prompt": "### Instruction:\n", + "ai_prompt": "### Response:\n" + }, + "parameters": { + "max_tokens": 2048 + }, + "metadata": { + "author": "Deepseek, The Bloke", + "tags": ["Community Recommended", "Code", "Big Context Length"], + "size": 26040000000 + } + } + \ No newline at end of file diff --git a/models/wizardcoder-34b/model.json b/models/wizardcoder-34b/model.json deleted file mode 100644 index ca4a3da8e..000000000 --- a/models/wizardcoder-34b/model.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "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 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", - "settings": { - "ctx_len": 2048, - "system_prompt": "", - "user_prompt": "### Instruction:\n", - "ai_prompt": "### Response:\n" - }, - "parameters": { - "max_tokens": 2048 - }, - "metadata": { - "author": "WizardLM, The Bloke", - "tags": ["Community Recommended", "Code", "Big Context Length"], - "size": 24320000000 - } - } - \ No newline at end of file From 69af5a034ad0026055259cd93b970107635a1a00 Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 16:38:13 +0700 Subject: [PATCH 15/17] update tags --- models/capybara-34b/model.json | 2 +- models/deepseek-coder-1.3b/model.json | 2 +- models/deepseek-coder-34b/model.json | 2 +- models/neural-chat-7b/model.json | 2 +- models/noromaid-20b/model.json | 2 +- models/openhermes-neural-7b/model.json | 2 +- models/openorca-13b/model.json | 2 +- models/phind-34b/model.json | 2 +- models/rocket-3b/model.json | 2 +- models/starling-7b/model.json | 2 +- models/tiefighter-13b/model.json | 2 +- models/tinyllama-1.1b/model.json | 2 +- models/wizardcoder-13b/model.json | 2 +- models/yi-34b/model.json | 2 +- models/zephyr-beta-7b/model.json | 2 +- 15 files changed, 15 insertions(+), 15 deletions(-) diff --git a/models/capybara-34b/model.json b/models/capybara-34b/model.json index e5834a66d..5f0bd7948 100644 --- a/models/capybara-34b/model.json +++ b/models/capybara-34b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "NousResearch, The Bloke", - "tags": ["Community Recommended", "General Use", "Big Context Length"], + "tags": ["Community Recommended", "Full Capabilities", "Finetuned"], "size": 24320000000 } } diff --git a/models/deepseek-coder-1.3b/model.json b/models/deepseek-coder-1.3b/model.json index 2bb9e7fee..47057626b 100644 --- a/models/deepseek-coder-1.3b/model.json +++ b/models/deepseek-coder-1.3b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Deepseek, The Bloke", - "tags": ["Community Recommended", "Code", "Small size"], + "tags": ["Foundational Model", "Code", "Small size"], "size": 870000000 } } diff --git a/models/deepseek-coder-34b/model.json b/models/deepseek-coder-34b/model.json index e175396f5..098cf87fd 100644 --- a/models/deepseek-coder-34b/model.json +++ b/models/deepseek-coder-34b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Deepseek, The Bloke", - "tags": ["Community Recommended", "Code", "Big Context Length"], + "tags": ["Foundational Model", "Community Recommended", "Code", "Big Context Length"], "size": 26040000000 } } diff --git a/models/neural-chat-7b/model.json b/models/neural-chat-7b/model.json index 234a2ff80..5357d8822 100644 --- a/models/neural-chat-7b/model.json +++ b/models/neural-chat-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Intel, The Bloke", - "tags": ["Community Recommended", "General Use", "Role-playing", "Big Context Length"], + "tags": ["Community Recommended", "Full Capabilities", "Big Context Length", "Finetuned"], "size": 4370000000 } } diff --git a/models/noromaid-20b/model.json b/models/noromaid-20b/model.json index da69cba3e..e8e3a9128 100644 --- a/models/noromaid-20b/model.json +++ b/models/noromaid-20b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "NeverSleep, The Bloke", - "tags": ["Community Recommended", "Role-playing"], + "tags": ["Community Recommended", "Role-playing", "Finetuned"], "size": 12040000000 } } diff --git a/models/openhermes-neural-7b/model.json b/models/openhermes-neural-7b/model.json index a7366f85b..2e9452067 100644 --- a/models/openhermes-neural-7b/model.json +++ b/models/openhermes-neural-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Intel, The Bloke", - "tags": ["Community Recommended", "General Use", "Code", "Merged"], + "tags": ["Community Recommended", "Full Capabilities", "Merged"], "size": 4370000000 } } diff --git a/models/openorca-13b/model.json b/models/openorca-13b/model.json index edfef0807..709cb0d36 100644 --- a/models/openorca-13b/model.json +++ b/models/openorca-13b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Microsoft, The Bloke", - "tags": ["Community Recommended", "General Use"], + "tags": ["Community Recommended", "General", "Reasoning", "Finetuned"], "size": 9230000000 } } diff --git a/models/phind-34b/model.json b/models/phind-34b/model.json index 8de7a84d0..532ad70fe 100644 --- a/models/phind-34b/model.json +++ b/models/phind-34b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Phind, The Bloke", - "tags": ["Community Recommended", "Code", "Big Context Length"], + "tags": ["Community Recommended", "Code", "Big Context Length", "Finetuned"], "size": 24320000000 } } diff --git a/models/rocket-3b/model.json b/models/rocket-3b/model.json index 7435bfafb..cd790d83c 100644 --- a/models/rocket-3b/model.json +++ b/models/rocket-3b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "pansophic, The Bloke", - "tags": ["Community Recommended", "General Use"], + "tags": ["Community Recommended", "General", "Small Size"], "size": 1710000000 } } \ No newline at end of file diff --git a/models/starling-7b/model.json b/models/starling-7b/model.json index a7ba963fe..5da638a29 100644 --- a/models/starling-7b/model.json +++ b/models/starling-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Berkeley-nest, The Bloke", - "tags": ["Community Recommended", "General", "Code"], + "tags": ["Community Recommended", "Full Capabilities","Finetuned"], "size": 4370000000 } } diff --git a/models/tiefighter-13b/model.json b/models/tiefighter-13b/model.json index 9940c9a70..e7b962e65 100644 --- a/models/tiefighter-13b/model.json +++ b/models/tiefighter-13b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "KoboldAI, The Bloke", - "tags": ["Community Recommended", "General Use", "Role-playing", "Writing"], + "tags": ["Community Recommended", "Role-playing", "Creative", "Finetuned"], "size": 9230000000 } } diff --git a/models/tinyllama-1.1b/model.json b/models/tinyllama-1.1b/model.json index 92793d7f6..93fd35fc6 100644 --- a/models/tinyllama-1.1b/model.json +++ b/models/tinyllama-1.1b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "TinyLlama", - "tags": ["Community Recommended", "General Use"], + "tags": ["Community Recommended", "General", "Small Size"], "size": 637000000 } } \ No newline at end of file diff --git a/models/wizardcoder-13b/model.json b/models/wizardcoder-13b/model.json index a7284aa7c..fc1f6f33c 100644 --- a/models/wizardcoder-13b/model.json +++ b/models/wizardcoder-13b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "WizardLM, The Bloke", - "tags": ["Community Recommended", "Code", "Big Context Length"], + "tags": ["Community Recommended", "Code", "Big Context Length", "Finetuned"], "size": 9230000000 } } diff --git a/models/yi-34b/model.json b/models/yi-34b/model.json index bc6e8e9e3..7e574daf0 100644 --- a/models/yi-34b/model.json +++ b/models/yi-34b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "01-ai, The Bloke", - "tags": ["Foundational Model", "General Use", "Role-playing", "Writing"], + "tags": ["Foundational Model", "Full Capabilities"], "size": 24320000000 } } diff --git a/models/zephyr-beta-7b/model.json b/models/zephyr-beta-7b/model.json index e1a175432..8b53ab07e 100644 --- a/models/zephyr-beta-7b/model.json +++ b/models/zephyr-beta-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "HuggingFaceH4, The Bloke", - "tags": ["Community Recommended", "General Use", "Big Context Length"], + "tags": ["Community Recommended", "Full Capabilities", "Finetuned"], "size": 4370000000 } } From 3f5317bb53de9e2c9599c3a0f2f514b85ab92660 Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 16:44:43 +0700 Subject: [PATCH 16/17] typo --- models/lzlv-70b/model.json | 2 +- models/rocket-3b/model.json | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/models/lzlv-70b/model.json b/models/lzlv-70b/model.json index e4247eb88..0b1577131 100644 --- a/models/lzlv-70b/model.json +++ b/models/lzlv-70b/model.json @@ -16,7 +16,7 @@ "max_tokens": 2048 }, "metadata": { - "author": "lizpreciatior, The Bloke", + "author": "Lizpreciatior, The Bloke", "tags": ["Community Recommended", "General Use", "Role-playing"], "size": 48750000000 } diff --git a/models/rocket-3b/model.json b/models/rocket-3b/model.json index cd790d83c..83a61584a 100644 --- a/models/rocket-3b/model.json +++ b/models/rocket-3b/model.json @@ -16,7 +16,7 @@ "max_tokens": 2048 }, "metadata": { - "author": "pansophic, The Bloke", + "author": "Pansophic, The Bloke", "tags": ["Community Recommended", "General", "Small Size"], "size": 1710000000 } From ff69e4fe51abe2dcc29d9b485c09f238cb1b73ee Mon Sep 17 00:00:00 2001 From: hahuyhoang411 Date: Mon, 4 Dec 2023 17:33:26 +0700 Subject: [PATCH 17/17] update models --- models/capybara-34b/model.json | 2 +- models/deepseek-coder-1.3b/model.json | 2 +- models/deepseek-coder-34b/model.json | 2 +- models/llama2-chat-70b-q4/model.json | 2 +- models/llama2-chat-7b-q4/model.json | 2 +- models/llama2-chat-7b-q5/model.json | 2 +- models/lzlv-70b/model.json | 2 +- models/mistral-ins-7b-q4/model.json | 2 +- models/mistral-ins-7b-q5/model.json | 2 +- models/neural-chat-7b/model.json | 2 +- models/noromaid-20b/model.json | 2 +- models/openhermes-neural-7b/model.json | 2 +- models/openorca-13b/model.json | 2 +- models/phind-34b/model.json | 2 +- models/rocket-3b/model.json | 2 +- models/starling-7b/model.json | 2 +- models/tiefighter-13b/model.json | 2 +- models/tinyllama-1.1b/model.json | 2 +- models/wizardcoder-13b/model.json | 2 +- models/yi-34b/model.json | 2 +- models/zephyr-beta-7b/model.json | 2 +- 21 files changed, 21 insertions(+), 21 deletions(-) diff --git a/models/capybara-34b/model.json b/models/capybara-34b/model.json index 5f0bd7948..14cdf1d04 100644 --- a/models/capybara-34b/model.json +++ b/models/capybara-34b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "NousResearch, The Bloke", - "tags": ["Community Recommended", "Full Capabilities", "Finetuned"], + "tags": ["Recommended", "Medium", "Finetuned"], "size": 24320000000 } } diff --git a/models/deepseek-coder-1.3b/model.json b/models/deepseek-coder-1.3b/model.json index 47057626b..ec2e2815f 100644 --- a/models/deepseek-coder-1.3b/model.json +++ b/models/deepseek-coder-1.3b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Deepseek, The Bloke", - "tags": ["Foundational Model", "Code", "Small size"], + "tags": ["Tiny", "Foundational Model"], "size": 870000000 } } diff --git a/models/deepseek-coder-34b/model.json b/models/deepseek-coder-34b/model.json index 098cf87fd..1936df1b9 100644 --- a/models/deepseek-coder-34b/model.json +++ b/models/deepseek-coder-34b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Deepseek, The Bloke", - "tags": ["Foundational Model", "Community Recommended", "Code", "Big Context Length"], + "tags": ["Recommended", "Medium", "Foundational Model"], "size": 26040000000 } } diff --git a/models/llama2-chat-70b-q4/model.json b/models/llama2-chat-70b-q4/model.json index 4ab3ac555..6d089f9fa 100644 --- a/models/llama2-chat-70b-q4/model.json +++ b/models/llama2-chat-70b-q4/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "MetaAI, The Bloke", - "tags": ["Foundational Model", "General", "Code"], + "tags": ["Large", "Foundational Model"], "size": 43920000000 } } diff --git a/models/llama2-chat-7b-q4/model.json b/models/llama2-chat-7b-q4/model.json index 7b52e1767..3ba054018 100644 --- a/models/llama2-chat-7b-q4/model.json +++ b/models/llama2-chat-7b-q4/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "MetaAI, The Bloke", - "tags": ["Foundational Model", "General", "Code"], + "tags": ["Small", "Foundational Model"], "size": 4080000000 } } diff --git a/models/llama2-chat-7b-q5/model.json b/models/llama2-chat-7b-q5/model.json index b19213e46..2889868d5 100644 --- a/models/llama2-chat-7b-q5/model.json +++ b/models/llama2-chat-7b-q5/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "MetaAI, The Bloke", - "tags": ["Foundational Model", "General", "Code"], + "tags": ["Small", "Foundational Model"], "size": 4780000000 } } diff --git a/models/lzlv-70b/model.json b/models/lzlv-70b/model.json index 0b1577131..d26ba7d29 100644 --- a/models/lzlv-70b/model.json +++ b/models/lzlv-70b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Lizpreciatior, The Bloke", - "tags": ["Community Recommended", "General Use", "Role-playing"], + "tags": ["Recommended", "Large", "Finetuned"], "size": 48750000000 } } diff --git a/models/mistral-ins-7b-q4/model.json b/models/mistral-ins-7b-q4/model.json index dd3360700..fe1bb0a5b 100644 --- a/models/mistral-ins-7b-q4/model.json +++ b/models/mistral-ins-7b-q4/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "MistralAI, The Bloke", - "tags": ["Foundational Model", "General", "Code"], + "tags": ["Small", "Foundational Model"], "size": 4370000000 } } diff --git a/models/mistral-ins-7b-q5/model.json b/models/mistral-ins-7b-q5/model.json index 5b86ab538..13eb6bde9 100644 --- a/models/mistral-ins-7b-q5/model.json +++ b/models/mistral-ins-7b-q5/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "MistralAI, The Bloke", - "tags": ["Foundational Model", "General", "Code"], + "tags": ["Small", "Foundational Model"], "size": 5130000000 } } diff --git a/models/neural-chat-7b/model.json b/models/neural-chat-7b/model.json index 5357d8822..cd0b3b3d0 100644 --- a/models/neural-chat-7b/model.json +++ b/models/neural-chat-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Intel, The Bloke", - "tags": ["Community Recommended", "Full Capabilities", "Big Context Length", "Finetuned"], + "tags": ["Recommended", "Small", "Finetuned"], "size": 4370000000 } } diff --git a/models/noromaid-20b/model.json b/models/noromaid-20b/model.json index e8e3a9128..f6994d1ce 100644 --- a/models/noromaid-20b/model.json +++ b/models/noromaid-20b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "NeverSleep, The Bloke", - "tags": ["Community Recommended", "Role-playing", "Finetuned"], + "tags": ["Medium", "Finetuned"], "size": 12040000000 } } diff --git a/models/openhermes-neural-7b/model.json b/models/openhermes-neural-7b/model.json index 2e9452067..536345fd1 100644 --- a/models/openhermes-neural-7b/model.json +++ b/models/openhermes-neural-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Intel, The Bloke", - "tags": ["Community Recommended", "Full Capabilities", "Merged"], + "tags": ["Recommended", "Small", "Merged"], "size": 4370000000 } } diff --git a/models/openorca-13b/model.json b/models/openorca-13b/model.json index 709cb0d36..0c0c9fff7 100644 --- a/models/openorca-13b/model.json +++ b/models/openorca-13b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Microsoft, The Bloke", - "tags": ["Community Recommended", "General", "Reasoning", "Finetuned"], + "tags": ["Medium", "Finetuned"], "size": 9230000000 } } diff --git a/models/phind-34b/model.json b/models/phind-34b/model.json index 532ad70fe..c8e2cb919 100644 --- a/models/phind-34b/model.json +++ b/models/phind-34b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Phind, The Bloke", - "tags": ["Community Recommended", "Code", "Big Context Length", "Finetuned"], + "tags": ["Recommended", "Medium", "Finetuned"], "size": 24320000000 } } diff --git a/models/rocket-3b/model.json b/models/rocket-3b/model.json index 83a61584a..c40ee6258 100644 --- a/models/rocket-3b/model.json +++ b/models/rocket-3b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Pansophic, The Bloke", - "tags": ["Community Recommended", "General", "Small Size"], + "tags": ["Tiny", "Finetuned"], "size": 1710000000 } } \ No newline at end of file diff --git a/models/starling-7b/model.json b/models/starling-7b/model.json index 5da638a29..1bc2aac61 100644 --- a/models/starling-7b/model.json +++ b/models/starling-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "Berkeley-nest, The Bloke", - "tags": ["Community Recommended", "Full Capabilities","Finetuned"], + "tags": ["Recommended", "Small","Finetuned"], "size": 4370000000 } } diff --git a/models/tiefighter-13b/model.json b/models/tiefighter-13b/model.json index e7b962e65..a28f88d85 100644 --- a/models/tiefighter-13b/model.json +++ b/models/tiefighter-13b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "KoboldAI, The Bloke", - "tags": ["Community Recommended", "Role-playing", "Creative", "Finetuned"], + "tags": ["Medium", "Finetuned"], "size": 9230000000 } } diff --git a/models/tinyllama-1.1b/model.json b/models/tinyllama-1.1b/model.json index 93fd35fc6..ace0ca6a0 100644 --- a/models/tinyllama-1.1b/model.json +++ b/models/tinyllama-1.1b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "TinyLlama", - "tags": ["Community Recommended", "General", "Small Size"], + "tags": ["Tiny", "Foundation Model"], "size": 637000000 } } \ No newline at end of file diff --git a/models/wizardcoder-13b/model.json b/models/wizardcoder-13b/model.json index fc1f6f33c..16f8946be 100644 --- a/models/wizardcoder-13b/model.json +++ b/models/wizardcoder-13b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "WizardLM, The Bloke", - "tags": ["Community Recommended", "Code", "Big Context Length", "Finetuned"], + "tags": ["Recommended", "Medium", "Finetuned"], "size": 9230000000 } } diff --git a/models/yi-34b/model.json b/models/yi-34b/model.json index 7e574daf0..c2d2fd327 100644 --- a/models/yi-34b/model.json +++ b/models/yi-34b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "01-ai, The Bloke", - "tags": ["Foundational Model", "Full Capabilities"], + "tags": ["Medium", "Foundational Model"], "size": 24320000000 } } diff --git a/models/zephyr-beta-7b/model.json b/models/zephyr-beta-7b/model.json index 8b53ab07e..c6b4bcdc7 100644 --- a/models/zephyr-beta-7b/model.json +++ b/models/zephyr-beta-7b/model.json @@ -17,7 +17,7 @@ }, "metadata": { "author": "HuggingFaceH4, The Bloke", - "tags": ["Community Recommended", "Full Capabilities", "Finetuned"], + "tags": ["Recommended", "Small", "Finetuned"], "size": 4370000000 } }