chore: sync initial hub models (#4778)
* chore: sync initial hub models * fix: openai request template
This commit is contained in:
Louis
GitHub
parent
3168ce7016
commit
b0deeed937
@ -1,7 +1,7 @@
|
||||
{
|
||||
"name": "@janhq/engine-management-extension",
|
||||
"productName": "Engine Management",
|
||||
"version": "1.0.0",
|
||||
"version": "1.0.1",
|
||||
"description": "Manages AI engines and their configurations.",
|
||||
"main": "dist/index.js",
|
||||
"node": "dist/node/index.cjs.js",
|
||||
|
||||
@ -10,7 +10,7 @@
|
||||
"transform_req": {
|
||||
"chat_completions": {
|
||||
"url": "https://api.openai.com/v1/chat/completions",
|
||||
"template": "{ {% set first = true %} {% for key, value in input_request %} {% if key == \"model\" or key == \"temperature\" or key == \"store\" or key == \"messages\" or key == \"stream\" or key == \"presence_penalty\" or key == \"metadata\" or key == \"frequency_penalty\" or key == \"tools\" or key == \"tool_choice\" or key == \"logprobs\" or key == \"top_logprobs\" or key == \"logit_bias\" or key == \"n\" or key == \"modalities\" or key == \"prediction\" or key == \"response_format\" or key == \"service_tier\" or key == \"seed\" or key == \"stream_options\" or key == \"top_p\" or key == \"parallel_tool_calls\" or key == \"user\" or key == \"max_tokens\" or ((input_request.model == \"o1\" or input_request.model == \"o1-preview\" or input_request.model == \"o1-mini\" or input_request.model == \"o3\" or input_request.model == \"o3-mini\") and (key == \"stop\")) %} {% if not first %} , {% endif %} {% if key == \"messages\" and (input_request.model == \"o1\" or input_request.model == \"o1-preview\" or input_request.model == \"o1-mini\") and input_request.messages.0.role == \"system\" %} \"messages\": [{% for message in input_request.messages %} {% if not loop.is_first %} { \"role\": \"{{ message.role }}\", \"content\": \"{{ message.content }}\" } {% if not loop.is_last %} , {% endif %} {% endif %} {% endfor %}] {% else if key == \"max_tokens\" and (input_request.model == \"o1\" or input_request.model == \"o1-preview\" or input_request.model == \"o1-mini\" or input_request.model == \"o3\" or input_request.model == \"o3-mini\") %} \"max_completion_tokens\": {{ tojson(value) }} {% set first = false %} {% else %} \"{{ key }}\": {{ tojson(value) }} {% set first = false %} {% endif %} {% endif %} {% endfor %} }"
|
||||
"template": "{ {% set first = true %} {% for key, value in input_request %} {% if key == \"model\" or key == \"temperature\" or key == \"store\" or key == \"messages\" or key == \"stream\" or key == \"presence_penalty\" or key == \"metadata\" or key == \"frequency_penalty\" or key == \"tools\" or key == \"tool_choice\" or key == \"logprobs\" or key == \"top_logprobs\" or key == \"logit_bias\" or key == \"n\" or key == \"modalities\" or key == \"prediction\" or key == \"response_format\" or key == \"service_tier\" or key == \"seed\" or key == \"stream_options\" or key == \"top_p\" or key == \"parallel_tool_calls\" or key == \"user\" or key == \"max_tokens\" or key == \"stop\" %} {% if not first %}, {% endif %} {% if key == \"messages\" and (input_request.model == \"o1\" or input_request.model == \"o1-preview\" or input_request.model == \"o1-mini\") and input_request.messages.0.role == \"system\" %} \"messages\": [ {% for message in input_request.messages %} {% if not loop.is_first %} { \"role\": \"{{ message.role }}\", \"content\": \"{{ message.content }}\" } {% if not loop.is_last %}, {% endif %} {% endif %} {% endfor %} ] {% else if key == \"stop\" and (input_request.model == \"o1\" or input_request.model == \"o1-preview\" or input_request.model == \"o1-mini\" or input_request.model == \"o3\" or input_request.model == \"o3-mini\") %} {% set first = false %} {% else if key == \"max_tokens\" and (input_request.model == \"o1\" or input_request.model == \"o1-preview\" or input_request.model == \"o1-mini\" or input_request.model == \"o3\" or input_request.model == \"o3-mini\") %} \"max_completion_tokens\": {{ tojson(value) }} {% set first = false %} {% else %} \"{{ key }}\": {{ tojson(value) }} {% set first = false %} {% endif %} {% endif %} {% endfor %} }"
|
||||
}
|
||||
},
|
||||
"transform_resp": {
|
||||
|
||||
@ -9,7 +9,7 @@
|
||||
"license": "cc-by-nc-4.0"
|
||||
},
|
||||
"createdAt": "2024-06-21T06:20:08.000Z",
|
||||
"description": "---\nlicense: cc-by-nc-4.0\n---\n\n## Overview\n\nC4AI Command-R is a research release of a 35 billion parameter highly performant generative model. Command-R is a large language model with open weights optimized for a variety of use cases including reasoning, summarization, and question answering. Command-R has the capability for multilingual generation evaluated in 10 languages and highly performant RAG capabilities.\n\n## Variants\n\n| No | Variant | Cortex CLI command |\n| --- | --- | --- |\n| 1 | [35b-gguf](https://huggingface.co/cortexhub/command-r/tree/35b-gguf) | `cortex run command-r:35b-gguf` |\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```\n cortexhub/command-r\n ```\n \n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```\n cortex run command-r\n ```\n \n## Credits\n\n- **Author:** Cohere For AI: [cohere.for.ai](https://cohere.for.ai/)\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [Licence](https://cohere.com/c4ai-cc-by-nc-license)",
|
||||
"description": "---\nlicense: cc-by-nc-4.0\npipeline_tag: text-generation\ntags:\n- cortex.cpp\n---\n\n## Overview\n\nC4AI Command-R is a research release of a 35 billion parameter highly performant generative model. Command-R is a large language model with open weights optimized for a variety of use cases including reasoning, summarization, and question answering. Command-R has the capability for multilingual generation evaluated in 10 languages and highly performant RAG capabilities.\n\n## Variants\n\n| No | Variant | Cortex CLI command |\n| --- | --- | --- |\n| 1 | [Command-r-32b](https://huggingface.co/cortexhub/command-r/tree/32b) | `cortex run command-r:32b` |\n| 1 | [Command-r-35b](https://huggingface.co/cortexhub/command-r/tree/35b) | `cortex run command-r:35b` |\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```bash\n cortexhub/command-r\n ```\n \n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```bash\n cortex run command-r\n ```\n \n## Credits\n\n- **Author:** Cohere For AI\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [Licence](https://cohere.com/c4ai-cc-by-nc-license)",
|
||||
"disabled": false,
|
||||
"downloads": 14,
|
||||
"gated": false,
|
||||
@ -99,7 +99,7 @@
|
||||
"license": "gemma"
|
||||
},
|
||||
"createdAt": "2024-08-05T06:07:51.000Z",
|
||||
"description": "---\nlicense: gemma\n---\n\n## Overview\n\nThe [Gemma](https://huggingface.co/google/gemma-2-2b-it), state-of-the-art open model trained with the Gemma datasets that includes both synthetic data and the filtered publicly available websites data with a focus on high-quality and reasoning dense properties. The model belongs to the Gemma family with the 4B, 7B version in two variants 8K and 128K which is the context length (in tokens) that it can support.\n\n## Variants\n\n| No | Variant | Cortex CLI command |\n| --- | --- | --- |\n| 1 | [2b-gguf](https://huggingface.co/cortexso/gemma2/tree/2b-gguf) | `cortex run gemma:2b-gguf` |\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```\n cortexso/gemma2\n ```\n \n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```\n cortex run gemma2\n ```\n \n## Credits\n\n- **Author:** Go\u200cogle\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [License](https://ai.google.dev/gemma/terms)\n- **Papers:** [Gemma Technical Report](https://arxiv.org/abs/2403.08295)",
|
||||
"description": "---\nlicense: gemma\npipeline_tag: text-generation\ntags:\n- cortex.cpp\n---\n\n## Overview\n\nThe [Gemma](https://huggingface.co/google/gemma-2-2b-it), state-of-the-art open model trained with the Gemma datasets that includes both synthetic data and the filtered publicly available websites data with a focus on high-quality and reasoning dense properties. The model belongs to the Gemma family with the 4B, 7B version in two variants 8K and 128K which is the context length (in tokens) that it can support.\n\n## Variants\n\n| No | Variant | Cortex CLI command |\n| --- | --- | --- |\n| 1 | [Gemma2-2b](https://huggingface.co/cortexso/gemma2/tree/2b) | `cortex run gemma2:2b` |\n| 2 | [Gemma2-9b](https://huggingface.co/cortexso/gemma2/tree/9b) | `cortex run gemma2:9b` |\n| 3 | [Gemma2-27b](https://huggingface.co/cortexso/gemma2/tree/27b) | `cortex run gemma2:27b` |\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```bash\n cortexso/gemma2\n ```\n \n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```bash\n cortex run gemma2\n ```\n \n## Credits\n\n- **Author:** Go\u200cogle\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [License](https://ai.google.dev/gemma/terms)\n- **Papers:** [Gemma Technical Report](https://arxiv.org/abs/2403.08295)",
|
||||
"disabled": false,
|
||||
"downloads": 190,
|
||||
"gated": false,
|
||||
@ -273,7 +273,7 @@
|
||||
"license": "apache-2.0"
|
||||
},
|
||||
"createdAt": "2024-06-24T06:11:45.000Z",
|
||||
"description": "---\nlicense: apache-2.0\n---\n\n## Overview\n\nThe Aya model is a massively multilingual generative language model that follows instructions in 101 languages.\n\n## Variants\n\n| No | Variant | Cortex CLI command |\n| --- | --- | --- |\n| 1 | [12.9b-gguf](https://huggingface.co/cortexhub/aya/tree/12.9b-gguf) | `cortex run aya:12.9b-gguf` |\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```\n cortexhub/aya\n ```\n\n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```\n cortex run aya\n ```\n\n## Credits\n\n- **Author:** [Cohere For AI](https://cohere.for.ai)\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)",
|
||||
"description": "---\nlicense: cc-by-sa-4.0\npipeline_tag: text-generation\ntags:\n- cortex.cpp\n---\n\n## Overview\n\nAya Expanse is an open-weight research release of a model with highly advanced multilingual capabilities. It focuses on pairing a highly performant pre-trained Command family of models with the result of a year\u2019s dedicated research from Cohere For AI, including data arbitrage, multilingual preference training, safety tuning, and model merging. The result is a powerful multilingual large language model serving 23 languages.\n\n## Variants\n\n| No | Variant | Cortex CLI command |\n| --- | --- | --- |\n| 1 | [main](https://huggingface.co/cortexso/aya-expanse/tree/main) | `cortex run aya-expanse` |\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```\n cortexso/aya-expanse\n ```\n\n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```\n cortex run aya-expanse\n ```\n\n## Credits\n\n- **Author:** CohereAI\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [License](https://cohere.com/c4ai-cc-by-nc-license)\n- **Papers:** [Aya Expanse Blog](https://cohere.com/blog/aya-expanse-connecting-our-world)",
|
||||
"disabled": false,
|
||||
"downloads": 25,
|
||||
"gated": false,
|
||||
@ -325,7 +325,7 @@
|
||||
"tags": ["cortex.cpp"]
|
||||
},
|
||||
"createdAt": "2024-10-26T15:40:05.000Z",
|
||||
"description": "---\nlicense: apache-2.0\npipeline_tag: text-generation\ntags:\n- cortex.cpp\n---\n\n## Overview\n\nQwen2.5 by Qwen is a family of model include various specialized models for coding and mathematics available in multiple sizes from 0.5B to 72B parameters\n\n## Variants\n\n| No | Variant | Cortex CLI command |\n| --- | --- | --- |\n| 1 | [main/default](https://huggingface.co/cortexso/qwen2.5/tree/main) | `cortex run qwen2.5` |\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```\n cortexso/qwen2.5\n ```\n\n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```\n cortex run qwen2.5\n ```\n\n## Credits\n\n- **Author:** Qwen\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [License Apache 2.0](https://www.apache.org/licenses/LICENSE-2.0)\n- **Papers:** [Qwen2.5 Blog](https://qwenlm.github.io/blog/qwen2.5/)",
|
||||
"description": "---\nlicense: apache-2.0\npipeline_tag: text-generation\ntags:\n- cortex.cpp\n---\n\n## Overview\n\nQwen2.5 by Qwen is a family of model include various specialized models for coding and mathematics available in multiple sizes from 0.5B to 72B parameters\n\n## Variants\n\n| No | Variant | Cortex CLI command |\n| --- | --- | --- |\n| 1 | [Qwen-2.5-0.5b](https://huggingface.co/cortexso/qwen2.5/tree/0.5b) | `cortex run qwen2.5:0.5b` |\n| 2 | [Qwen-2.5-1.5b](https://huggingface.co/cortexso/qwen2.5/tree/1.5b) | `cortex run qwen2.5:1.5b` |\n| 3 | [Qwen-2.5-3b](https://huggingface.co/cortexso/qwen2.5/tree/3b) | `cortex run qwen2.5:3b` |\n| 4 | [Qwen-2.5-7b](https://huggingface.co/cortexso/qwen2.5/tree/7b) | `cortex run qwen2.5:7b` |\n| 5 | [Qwen-2.5-14b](https://huggingface.co/cortexso/qwen2.5/tree/14b) | `cortex run qwen2.5:14b` |\n| 6 | [Qwen-2.5-32b](https://huggingface.co/cortexso/qwen2.5/tree/32b) | `cortex run qwen2.5:32b` |\n| 7 | [Qwen-2.5-72b](https://huggingface.co/cortexso/qwen2.5/tree/72b) | `cortex run qwen2.5:72b` |\n| 8 | [Qwen-2.5-coder-1.5b](https://huggingface.co/cortexso/qwen2.5/tree/coder-1.5b) | `cortex run qwen2.5:coder-1.5b` |\n| 9 | [Qwen-2.5-coder-7b](https://huggingface.co/cortexso/qwen2.5/tree/coder-7b) | `cortex run qwen2.5:coder-7b` |\n| 10 | [Qwen-2.5-math-1.5b](https://huggingface.co/cortexso/qwen2.5/tree/math-1.5b) | `cortex run qwen2.5:math-1.5b` |\n| 11 | [Qwen-2.5-math-7b](https://huggingface.co/cortexso/qwen2.5/tree/math-7b) | `cortex run qwen2.5:math-7b` |\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```\n cortexso/qwen2.5\n ```\n\n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```\n cortex run qwen2.5\n ```\n\n## Credits\n\n- **Author:** Qwen\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [License Apache 2.0](https://choosealicense.com/licenses/apache-2.0/)\n- **Papers:** [Qwen2.5 Blog](https://qwenlm.github.io/blog/qwen2.5/)",
|
||||
"disabled": false,
|
||||
"downloads": 2482,
|
||||
"gated": false,
|
||||
@ -751,7 +751,7 @@
|
||||
"tags": ["cortex.cpp", "featured"]
|
||||
},
|
||||
"createdAt": "2024-09-27T04:22:33.000Z",
|
||||
"description": "---\nlicense: llama3.2\npipeline_tag: text-generation\ntags:\n- cortex.cpp\n- featured\n---\n\n## Overview\n\nMeta developed and released the [Meta Llama 3.2](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct) family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 1B and 3B sizes (text in/text out). The Llama 3.2 instruction-tuned text only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks. They outperform many of the available open source and closed chat models on common industry benchmarks.\n\n## Variants\n\n| No | Variant | Cortex CLI command |\n| --- | --- | --- |\n| 2 | [gguf](https://huggingface.co/cortexso/llama3.2/tree/gguf) | `cortex run llama3.2:gguf` |\n| 3 | [main/default](https://huggingface.co/cortexso/llama3.2/tree/main) | `cortex run llama3.2` |\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```\n cortexso/llama3.2\n ```\n\n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```\n cortex run llama3.2\n ```\n\n## Credits\n\n- **Author:** Meta\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [License](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct/blob/main/LICENSE.txt)\n- **Papers:** [Llama-3.2 Blog](https://ai.meta.com/blog/llama-3-2-connect-2024-vision-edge-mobile-devices/)",
|
||||
"description": "---\nlicense: llama3.2\npipeline_tag: text-generation\ntags:\n- cortex.cpp\n- featured\n---\n\n## Overview\n\nMeta developed and released the [Meta Llama 3.2](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct) family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 1B and 3B sizes (text in/text out). The Llama 3.2 instruction-tuned text only models are optimized for multilingual dialogue use cases, including agentic retrieval and summarization tasks. They outperform many of the available open source and closed chat models on common industry benchmarks.\n\n## Variants\n\n| No | Variant | Cortex CLI command |\n| --- | --- | --- |\n| 2 | [gguf](https://huggingface.co/cortexso/llama3.2/tree/main) | `cortex run llama3.2` |\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```bash\n cortexso/llama3.2\n ```\n\n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```bash\n cortex run llama3.2\n ```\n\n## Credits\n\n- **Author:** Meta\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [License](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct/blob/main/LICENSE.txt)\n- **Papers:** [Llama-3.2 Blog](https://ai.meta.com/blog/llama-3-2-connect-2024-vision-edge-mobile-devices/)",
|
||||
"disabled": false,
|
||||
"downloads": 761,
|
||||
"gated": false,
|
||||
@ -899,7 +899,7 @@
|
||||
"tags": ["cortexp.cpp", "featured"]
|
||||
},
|
||||
"createdAt": "2025-02-03T12:56:17.000Z",
|
||||
"description": "---\nlicense: mit\npipeline_tag: text-generation\ntags:\n- cortexp.cpp\n- featured\n---\n\n## Overview\n\n**DeepSeek** developed and released the **DeepSeek-R1** series, featuring multiple model sizes fine-tuned for high-performance text generation. These models are optimized for dialogue, reasoning, and information-seeking tasks, providing a balance of efficiency and accuracy while maintaining a smaller footprint compared to their original counterparts.\n\nThe DeepSeek-R1 models include distilled and full-scale variants of both **Qwen** and **Llama** architectures, catering to various applications such as customer support, conversational AI, research, and enterprise automation.\n\n## Variants\n\n### DeepSeek-R1\n\n| No | Variant | Branch | Cortex CLI command |\n| -- | ---------------------------------------------------------------------------------------------- | ------- | ------------------------------------------ |\n| 1 | [DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/cortexso/deepseek-r1/tree/1.5b) | 1.5b | `cortex run [WIP]` |\n| 2 | [DeepSeek-R1-Distill-Qwen-7B](https://huggingface.co/cortexso/deepseek-r1/tree/7b) | 7b | `cortex run [WIP]` |\n| 3 | [DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/cortexso/deepseek-r1/tree/8b) | 8b | `cortex run [WIP]` |\n| 4 | [DeepSeek-R1-Distill-Qwen-14B](https://huggingface.co/cortexso/deepseek-r1/tree/14b) | 14b | `cortex run [WIP]` |\n| 5 | [DeepSeek-R1-Distill-Qwen-32B](https://huggingface.co/cortexso/deepseek-r1/tree/32b) | 32b | `cortex run [WIP]` |\n| 6 | [DeepSeek-R1-Distill-Llama-70B](https://huggingface.co/cortexso/deepseek-r1/tree/70b) | 70b | `cortex run [WIP]` |\n\nEach branch contains a default quantized version:\n- **Qwen-1.5B:** q4-km\n- **Qwen-7B:** q4-km\n- **Llama-8B:** q4-km\n- **Qwen-14B:** q4-km\n- **Qwen-32B:** q4-km\n- **Llama-70B:** q4-km\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```text\n cortexso/deepseek-r1 [WIP]\n cortexso/deepseek-r1 [WIP]\n ```\n\n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```bash\n cortex run [WIP]\n ```\n or\n ```bash\n cortex run [WIP]\n ```\n\n## Credits\n\n- **Author:** DeepSeek\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [License](https://huggingface.co/deepseek-ai/DeepSeek-R1#license)\n- **Papers:** [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://arxiv.org/html/2501.12948v1)",
|
||||
"description": "---\nlicense: mit\npipeline_tag: text-generation\ntags:\n- cortexp.cpp\n- featured\n---\n\n## Overview\n\n**DeepSeek** developed and released the **DeepSeek-R1** series, featuring multiple model sizes fine-tuned for high-performance text generation. These models are optimized for dialogue, reasoning, and information-seeking tasks, providing a balance of efficiency and accuracy while maintaining a smaller footprint compared to their original counterparts.\n\nThe DeepSeek-R1 models include distilled and full-scale variants of both **Qwen** and **Llama** architectures, catering to various applications such as customer support, conversational AI, research, and enterprise automation.\n\n## Variants\n\n### DeepSeek-R1\n\n| No | Variant | Branch | Cortex CLI command |\n| -- | ---------------------------------------------------------------------------------------------- | ------- | ------------------------------------------ |\n| 1 | [DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/cortexso/deepseek-r1/tree/1.5b) | 1.5b | `cortex run deepseek-r1:1.5b` |\n| 2 | [DeepSeek-R1-Distill-Qwen-7B](https://huggingface.co/cortexso/deepseek-r1/tree/7b) | 7b | `cortex run deepseek-r1:7b` |\n| 3 | [DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/cortexso/deepseek-r1/tree/8b) | 8b | `cortex run deepseek-r1:8b` |\n| 4 | [DeepSeek-R1-Distill-Qwen-14B](https://huggingface.co/cortexso/deepseek-r1/tree/14b) | 14b | `cortex run deepseek-r1:14b` |\n| 5 | [DeepSeek-R1-Distill-Qwen-32B](https://huggingface.co/cortexso/deepseek-r1/tree/32b) | 32b | `cortex run deepseek-r1:32b` |\n| 6 | [DeepSeek-R1-Distill-Llama-70B](https://huggingface.co/cortexso/deepseek-r1/tree/70b) | 70b | `cortex run deepseek-r1:70b` |\n\nEach branch contains a default quantized version:\n- **Qwen-1.5B:** q4-km\n- **Qwen-7B:** q4-km\n- **Llama-8B:** q4-km\n- **Qwen-14B:** q4-km\n- **Qwen-32B:** q4-km\n- **Llama-70B:** q4-km\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```text\n cortexso/deepseek-r1\n ```\n\n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```bash\n cortex run deepseek-r1\n ```\n\n## Credits\n\n- **Author:** DeepSeek\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [License](https://huggingface.co/deepseek-ai/DeepSeek-R1#license)\n- **Papers:** [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://arxiv.org/html/2501.12948v1)",
|
||||
"disabled": false,
|
||||
"downloads": 999,
|
||||
"gated": false,
|
||||
@ -1153,7 +1153,7 @@
|
||||
"tags": ["cortex.cpp"]
|
||||
},
|
||||
"createdAt": "2024-07-29T10:25:05.000Z",
|
||||
"description": "---\nlicense: llama3.1\npipeline_tag: text-generation\ntags:\n- cortex.cpp\n---\n\n## Overview\n\nMeta developed and released the [Meta Llama 3.1](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 8 and 70B sizes. The Llama 3 instruction tuned models are optimized for dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Further, in developing these models, we took great care to optimize helpfulness and safety.\n\n## Variants\n\n| No | Variant | Cortex CLI command |\n| --- | --- | --- |\n| 2 | [gguf](https://huggingface.co/cortexso/llama3.1/tree/gguf) | `cortex run llama3.1:gguf` |\n| 3 | [main/default](https://huggingface.co/cortexso/llama3.1/tree/main) | `cortex run llama3.1` |\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```\n cortexso/llama3.1\n ```\n\n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```\n cortex run llama3.1\n ```\n\n## Credits\n\n- **Author:** Meta\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [License](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B/blob/main/LICENSE)\n- **Papers:** [Llama-3.1 Blog](https://scontent.fsgn3-1.fna.fbcdn.net/v/t39.2365-6/452387774_1036916434819166_4173978747091533306_n.pdf?_nc_cat=104&ccb=1-7&_nc_sid=3c67a6&_nc_ohc=DTS7hDTcxZoQ7kNvgHxaQ8K&_nc_ht=scontent.fsgn3-1.fna&oh=00_AYC1gXduoxatzt8eFMfLunrRUzpzQcoKzAktIOT7FieZAQ&oe=66AE9C4D)",
|
||||
"description": "---\nlicense: llama3.1\npipeline_tag: text-generation\ntags:\n- cortex.cpp\n---\n\n## Overview\n\nMeta developed and released the [Meta Llama 3.1](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) family of large language models (LLMs), a collection of pretrained and instruction tuned generative text models in 8 and 70B sizes. The Llama 3 instruction tuned models are optimized for dialogue use cases and outperform many of the available open source chat models on common industry benchmarks. Further, in developing these models, we took great care to optimize helpfulness and safety.\n\n## Variants\n\n| No | Variant | Cortex CLI command |\n| --- | --- | --- |\n| 1 | [Llama3.1-8b](https://huggingface.co/cortexso/llama3.1/tree/8b) | `cortex run llama3.1:8b` |\n\n## Use it with Jan (UI)\n\n1. Install **Jan** using [Quickstart](https://jan.ai/docs/quickstart)\n2. Use in Jan model Hub:\n ```bash\n cortexso/llama3.1\n ```\n\n## Use it with Cortex (CLI)\n\n1. Install **Cortex** using [Quickstart](https://cortex.jan.ai/docs/quickstart)\n2. Run the model with command:\n ```bash\n cortex run llama3.1\n ```\n\n## Credits\n\n- **Author:** Meta\n- **Converter:** [Homebrew](https://www.homebrew.ltd/)\n- **Original License:** [License](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B/blob/main/LICENSE)\n- **Papers:** [Llama-3.1 Blog](https://ai.meta.com/blog/meta-llama-3-1/)",
|
||||
"disabled": false,
|
||||
"downloads": 275,
|
||||
"gated": false,
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user