{"id":16883962,"url":"https://github.com/xyproto/llm-manager","last_synced_at":"2025-04-11T12:09:25.396Z","repository":{"id":257823155,"uuid":"865937614","full_name":"xyproto/llm-manager","owner":"xyproto","description":"Select LLM models for code completion, image recognition and text generation","archived":false,"fork":false,"pushed_at":"2024-11-06T12:51:35.000Z","size":45,"stargazers_count":13,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"main","last_synced_at":"2025-03-25T08:23:47.768Z","etag":null,"topics":["ollama","system-configuration","tool"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"bsd-3-clause","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/xyproto.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-10-01T11:26:53.000Z","updated_at":"2025-03-16T16:42:33.000Z","dependencies_parsed_at":"2024-10-12T17:06:58.301Z","dependency_job_id":null,"html_url":"https://github.com/xyproto/llm-manager","commit_stats":null,"previous_names":["xyproto/llm-manager"],"tags_count":5,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/xyproto%2Fllm-manager","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/xyproto%2Fllm-manager/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/xyproto%2Fllm-manager/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/xyproto%2Fllm-manager/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/xyproto","download_url":"https://codeload.github.com/xyproto/llm-manager/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248397644,"owners_count":21097144,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ollama","system-configuration","tool"],"created_at":"2024-10-13T16:15:56.476Z","updated_at":"2025-04-11T12:09:25.372Z","avatar_url":"https://github.com/xyproto.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# llm-manager\n\nA utility written in Python 3 for querying and managing keys and values in `~/.config/llm-manager/llm.conf` and `/etc/llm.conf`.\n\n**The idea is that if ie. an editor needs to use the best available LLM model for code completion, it can use this utility to get the user-configured models.**\n\n```bash\n$ llm-manager code-completion\ndeepseek-coder-v2:236b\n```\n\nThis way, the user can install and configure the best Ollama model that their laptop and/or Ollama server (defined with `OLLAMA_HOST`) can support.\n\n## Example use\n\nSet the default `text-generation` model to `gemma2:2b`:\n\n```bash\n$ llm-manager set text-generation gemma2:2b\n```\n\nGet the current `text-generation` model\n\n```bash\n$ llm-manager get text-generation\ngamma2:2b\n```\n\nA shortcut:\n\n```bash\n$ llm-manager text-generation\ngemma2:2b\n```\n\nChanging the `text-generation` model to `llama3.2` and the `3b` tag:\n\n```bash\n$ llm-manager set text-generation llama3.2:3b\n```\n\nGet the current `text-generation` model:\n\n```bash\n$ llm-manager text-generation\nllama3.2:3b\n```\n\n## Default values\n\nThe default values for the current version of `llm-manager` and `/etc/llm.conf` are:\n\n| Task            | Model               |\n|-----------------|---------------------|\n| chat            | llama3.2:3b         |\n| code            | deepseek-coder:1.3b |\n| code-completion | deepseek-coder:1.3b |\n| test            | tinyllama:1b        |\n| text-generation | gemma2:2b           |\n| tool-use        | llama3.2:3b         |\n| translation     | mixtral:8x7b        |\n| vision          | llava:7b            |\n\nThe default configuration may change over time as better models become available.\n\nFor now, relatively small models are chosen, so that more people can use them, even without a GPU.\n\nHere is the default configuration file, `llm.conf`:\n\n```configuration\n# For chatting\nchat=llama3.2:3b\n\n# For analyzing or generating code\ncode=deepseek-coder-v2:latest\n\n# For code completion / tab autocompletion\ncode-completion=deepseek-coder:1.3b\n\n# A small model, for quick tests\ntest=tinyllama:1b\n\n# Text generation\ntext-generation=gemma2:2b\n\n# Tool use and function calling\ntool-use=llama3.2:3b\n\n# For translating text (not single words, though)\ntranslation=mixtral:8x7b\n\n# Vision and image detection\nvision=llava:7b\n```\n\n## Related projects\n\n* The [usermodel](https://github.com/xyproto/usermodel) package, for Go.\n\n## General info\n\n* Version: 1.2.1\n* License: BSD-3\n* Author: Alexander F. Rødseth \u0026lt;xyproto@archlinux.org\u0026gt;\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fxyproto%2Fllm-manager","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fxyproto%2Fllm-manager","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fxyproto%2Fllm-manager/lists"}