{"id":26189570,"url":"https://github.com/zack4dev/llm--raycast_g4f","last_synced_at":"2025-03-12T00:39:12.776Z","repository":{"id":277024886,"uuid":"924253417","full_name":"Zack4DEV/llm--Raycast_g4f","owner":"Zack4DEV","description":"Raycast extension by @XInTheDark to use GPT, Claude, Llama, and more... all for FREE! + Full support for custom APIs.","archived":false,"fork":false,"pushed_at":"2025-02-11T17:46:33.000Z","size":14568,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-02-11T18:44:51.510Z","etag":null,"topics":["ai","api","gpt","nexra"],"latest_commit_sha":null,"homepage":"","language":"JavaScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Zack4DEV.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE.txt","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":"2025-01-29T17:22:40.000Z","updated_at":"2025-02-11T17:46:38.000Z","dependencies_parsed_at":"2025-02-11T18:55:13.119Z","dependency_job_id":null,"html_url":"https://github.com/Zack4DEV/llm--Raycast_g4f","commit_stats":null,"previous_names":["zack4dev/llm--raycast_g4f"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Zack4DEV%2Fllm--Raycast_g4f","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Zack4DEV%2Fllm--Raycast_g4f/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Zack4DEV%2Fllm--Raycast_g4f/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Zack4DEV%2Fllm--Raycast_g4f/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Zack4DEV","download_url":"https://codeload.github.com/Zack4DEV/llm--Raycast_g4f/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":243136244,"owners_count":20241988,"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":["ai","api","gpt","nexra"],"created_at":"2025-03-12T00:39:12.221Z","updated_at":"2025-03-12T00:39:12.762Z","avatar_url":"https://github.com/Zack4DEV.png","language":"JavaScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Raycast-G4F (GPT4Free)\n\n## About\n\n**Powerful GPT-4, Llama-3 and more AI models on Raycast, for FREE! + Full support for custom APIs.**\n\nHomepage · [Privacy Policy](Privacy.md) · [FAQs](#FAQs) · [Discord][discord-invite]\n\n## Screenshots\n\n\u003cimg src=https://github.com/user-attachments/assets/1085b901-c72f-4bd9-9a37-473057e95393 width=400 alt=\"1\"\u003e\n\u003cimg src=https://github.com/user-attachments/assets/fb74f124-f1ac-4957-928e-363394ac0c3b width=400 alt=\"2\"\u003e\n\u003cimg src=https://github.com/user-attachments/assets/cf425e1a-8eaa-49d1-af31-e5fe8c6702db width=400 alt=\"3\"\u003e\n\u003cimg src=https://github.com/user-attachments/assets/7ab0da34-8f4d-4414-9548-901336e85588 width=400 alt=\"4\"\u003e\n\n## Workspace\n\n\"a ✨star✨ tysm!\" giving support the developer, probably\n\n#### Prerequisites\n\n- Install the [Raycast app](https://raycast.com/).\n - Currently, Raycast is only available for macOS. Windows support is [in progress](https://www.raycast.com/windows).\n- Install [Node.js](https://nodejs.org/).\n - The minimum version recommended is `v20.18.1`. An older version, should encounter issues.\n\n#### Installation\n\nThis extension is currently not available on the Raycast Extension store, but installation from source is extremely simple.\n\n1. Download the source code from [the latest release](https://github.com/XInTheDark/raycast-g4f/releases/latest), or\n clone the repository.\n2. Navigate to the directory, and open a Terminal window at the downloaded folder.\n3. Run `npm ci --production` to install required dependencies.\n4. (Optional) Run `pip3 install -r requirements.txt` to install Python dependencies. These are required for\n some features, e.g. web search.\n5. Run `npm run dev` to build and import the extension.\n\nThe extension, and its full set of commands, should then show up in your Raycast app.\n\n### Troubleshooting\n\nOpen an issue if any unexpected problems occur during installation.\n\n## Updating\n\n#### Automatically\n\nThere is built-in support for updating within the extension itself! Simply run the \"Check for Updates\" command in the extension, and it will take care of the update process for you. Furthermore, the \"Automatically Check for Updates\" feature is available in the preferences (enabled by default).\n\n#### Manually\n\nIn the command line, run `git pull`, `npm ci --production` and `npm run dev` (in that order).\n\nYou might want to update manually if the automatic update doesn't work (Open a GitHub issue if this is the case);\nupdating manually also allows you to fetch and view the latest changes to the source code.\n\n## Summary\n\n- ▶️️ Streaming support - see messages load in real-time, providing a seamless experience.\n- ⚡ Ask anything from anywhere - with 18 commands available, there's something for you no matter what you need.\n- 💪 Support for many providers \u0026 models (more info below!)\n- 💬 Chat command - interact with the AI in a conversation, and your chat history will be stored in the extension.\n- 🌐 Web search - let GPT search the web for the latest information.\n- 📄 File upload - you can upload image, video, audio and text files to the AI. (only available for a few providers, more\n to come!)\n- 🎨 Image generation capabilities - imagine anything, and make it reality with state-of-the-art models.\n- ✏️ Custom AI Commands - create your own commands with custom prompts!\n\n## Providers \u0026 Models\n\n\u003cdetails\u003e\n\n\u003csummary\u003eList of built-in providers\u003c/summary\u003e\n\n*Note: May not be up-to-date; please refer to the extension.*\n\n| Provider | Model | Features | Status | Speed | Rating and remarks by extension author |\n|------------------------------|-----------------------------------------|----------|-----------------------------|----------------|---------------------------------------------------------------------------------------------|\n| Nexra | gpt-4o (default) | ▶️ | ![Active][active-badge] | Very fast | 8.5/10, the best performing model. |\n| Nexra | gpt-4-32k | | ![Active][active-badge] | Medium | 6.5/10, no streaming support but otherwise a great model. |\n| Nexra | chatgpt | ▶️ | ![Unknown][unknown-badge] | Very fast | 7.5/10 |\n| Nexra | Bing | ▶️ | ![Active][active-badge] | Medium | 8/10, GPT-4 based with web search capabilities. |\n| Nexra | llama-3.1 | ▶️ | ![Active][active-badge] | Fast | 7/10 |\n| Nexra | gemini-1.0-pro | ▶️ | ![Active][active-badge] | Fast | 6.5/10 |\n| DeepInfra | meta-llama-3.3-70b | ▶️ | ![Active][active-badge] | Fast | 8.5/10, recent model with large context size. |\n| DeepInfra | meta-llama-3.2-90b-vision | ▶️ 📄¹ | ![Active][active-badge] | Fast | 8/10, recent model with vision capabilities. |\n| DeepInfra | meta-llama-3.2-11b-vision | ▶️ 📄¹ | ![Active][active-badge] | Very fast | 7.5/10 |\n| DeepInfra | meta-llama-3.1-405b | ▶️ | ![Inactive][inactive-badge] | Medium | 8.5/10, state-of-the-art open model, suitable for complex tasks. |\n| DeepInfra | meta-llama-3.1-70b | ▶️ | ![Active][active-badge] | Fast | 8/10 |\n| DeepInfra | meta-llama-3.1-8b | ▶️ | ![Active][active-badge] | Very fast | 7.5/10 |\n| DeepInfra | llama-3.1-nemotron-70b | ▶️ | ![Active][active-badge] | Fast | 8/10 |\n| DeepInfra | WizardLM-2-8x22B | ▶️ | ![Active][active-badge] | Medium | 7/10 |\n| DeepInfra | DeepSeek-V2.5 | ▶️ | ![Active][active-badge] | Fast | 7.5/10 |\n| DeepInfra | Qwen2.5-72B | ▶️ | ![Active][active-badge] | Medium | 7.5/10 |\n| DeepInfra | Qwen2.5-Coder-32B | ▶️ | ![Active][active-badge] | Fast | 7/10 |\n| DeepInfra | QwQ-32B-Preview | ▶️ | ![Active][active-badge] | Very fast | 7.5/10 |\n| Blackbox | custom model | ▶️ | ![Active][active-badge] | Fast | 7.5/10, very fast generation with built-in web search ability, but is optimized for coding. |\n| Blackbox | llama-3.1-405b | ▶️ | ![Active][active-badge] | Fast | 8.5/10 |\n| Blackbox | llama-3.1-70b | ▶️ | ![Active][active-badge] | Very fast | 8/10 |\n| Blackbox | llama-3.3-70b | ▶️ | ![Active][active-badge] | Very fast | 8/10 |\n| Blackbox | gemini-1.5-flash | ▶️ | ![Active][active-badge] | Extremely fast | 7.5/10 |\n| Blackbox | qwq-32b-preview | ▶️ | ![Active][active-badge] | Extremely fast | 6.5/10 |\n| Blackbox | gpt-4o | ▶️ | ![Active][active-badge] | Very fast | 7.5/10 |\n| Blackbox | claude-3.5-sonnet | ▶️ | ![Active][active-badge] | Fast | 8.5/10 |\n| Blackbox | gemini-pro | ▶️ | ![Active][active-badge] | Fast | 8/10 |\n| DuckDuckGo | gpt-4o-mini | ▶️ | ![Active][active-badge] | Extremely fast | 8/10, authentic GPT-4o-mini model with strong privacy. |\n| DuckDuckGo | claude-3-haiku | ▶️️ | ![Active][active-badge] | Extremely fast | 7/10 |\n| DuckDuckGo | meta-llama-3.1-70b | ▶️️ | ![Active][active-badge] | Very fast | 7.5/10 |\n| DuckDuckGo | mixtral-8x7b | ▶️️ | ![Active][active-badge] | Extremely fast | 7.5/10 |\n| BestIM | gpt-4o-mini | ▶️ | ![Inactive][inactive-badge] | Extremely fast | 8.5/10 |\n| AI4Chat | gpt-4 | | ![Active][active-badge] | Very fast | 7.5/10 |\n| DarkAI | gpt-4o | ▶️ | ![Active][active-badge] | Very fast | 8/10 |\n| Mhystical | gpt-4-32k | | ![Active][active-badge] | Very fast | 6.5/10 |\n| PizzaGPT | gpt-4o-mini | | ![Active][active-badge] | Extremely fast | 7.5/10 |\n| Meta AI | meta-llama-3.1 | ▶️ | ![Active][active-badge] | Medium | 7/10, recent model with internet access. |\n| Replicate | mixtral-8x7b | ▶️ | ![Active][active-badge] | Medium | ?/10 |\n| Replicate | meta-llama-3.1-405b | ▶️ | ![Active][active-badge] | Medium | ?/10 |\n| Replicate | meta-llama-3-70b | ▶️ | ![Active][active-badge] | Medium | ?/10 |\n| Replicate | meta-llama-3-8b | ▶️ | ![Active][active-badge] | Fast | ?/10 |\n| Phind | Phind Instant | ▶️ | ![Active][active-badge] | Extremely fast | 8/10 |\n| Google Gemini | auto (gemini-1.5-pro, gemini-1.5-flash) | ▶️ 📄 | ![Active][active-badge] | Very fast | 9/10, very good overall model but requires an API Key. (It's *free*, see the section below) |\n| Google Gemini (Experimental) | auto (changes frequently) | ▶️ 📄 | ![Active][active-badge] | Very fast | - |\n| Google Gemini (Thinking) | auto (changes frequently) | ▶️ 📄 | ![Active][active-badge] | Very fast | - |\n| Custom OpenAI-compatible API | - | ▶️ | ![Active][active-badge] | - | allows you to use any custom OpenAI-compatible API. [read more][local-api-help] |\n\n▶️ - Supports streaming.\n\n📄 - Supports file upload.\n**Note**: By default, all providers support basic file upload functionality for text-based files, like .txt, .md, etc.\n\n*¹: Supports images only.*\n\n\u003c/details\u003e\n\n#### Custom APIs\n\nThe extension fully supports custom APIs! You can use any OpenAI-compatible API and connect it.\n\nGet started by reading the [help page][local-api-help].\n\n#### Provider-specific notes\n\n- **Google Gemini**: An API Key is required to use this model. You can get one *completely for free*:\n\n1. Go to https://aistudio.google.com/app/apikey\n2. Sign in to your Google account if you haven't done so.\n3. Click on \"Create API Key\" and follow the instructions there.\n4. Copy the API Key and paste it into the corresponding box in the extension preferences.\n\nThe rate limit for Google Gemini is 1500 requests per day (as of the time of writing). This should be much more than\nenough for any normal usage.\nIf your use case needs an increased rate limit, you can even create multiple API Keys with different Google accounts;\nseparate them with commas in the preferences.\n\n- **Google Gemini**: This provider supports **File upload** functionality, as well as the **Ask About Screen Content**\n command!\n To upload a file in AI Chat, press Command-Enter or select \"Compose Message\" from the actions. Then, simply click on\n the upload button to get started.\n\n## Features\n\n\u003e [!NOTE]\n\u003e As of v5.0, the extension preferences are now found in the \"Preferences\" command. Please use this command to access\n\u003e the preferences instead of the Raycast preferences.\n\n#### Web Search\n\nLet GPT decide to search the web for information if it does not have enough knowledge or context. Uses DuckDuckGo\nsearch, fast and free.\n\n#### Usage\n\nEnabling web search is fast and easy. Go to the extension preferences, and the \"Web Search\" option will be available.\nThere are 4 options:\n\n- Disabled (default)\n- Automatic: Enable Web Search only in AI Chat. GPT will automatically decide when to use it.\n- Balanced: Use Web Search in every query for AI commands¹, and automatically in AI Chat. This is basically an extension\n of the \"Automatic\" option.\n- Always: Always use Web Search for every query, both in AI Chat and in commands¹.\n\n*¹: Commands that support Web Search are: Ask AI, Ask About Selected Text, Explain. Other commands will not use Web\nSearch.*\n\nWeb Search is also available in the following commands:\n\n- Custom AI Commands: You can enable Web Search for each command individually.\n- AI Chat: You can enable Web Search for each chat individually.\n- AI Presets: You can enable Web Search for each preset individually.\n\n#### Smart Chat Naming\n\nLet GPT automatically come up with a name for the current chat session after you send the first message. For example,\nthis is similar to what the ChatGPT web UI does.\n\n#### Automatically Check for Updates\n\nLet the extension automatically check for updates every day. If a new version is available, you will be notified,\nalong with the option to update the extension with a single click.\n\n#### Persistent Storage\n\nEnable more persistent storage of the extension's data, like AI Chat data or Custom Commands.\nThis will back up a copy of this data to files on your computer. Useful for saving large amounts of data.\n*Note: With this option off, your data is already well preserved. Do not enable this if you have sensitive data.*\n\n#### Enable Cursor Icon\n\nShow a cursor icon when the response is loading - cosmetic option only.\n\n#### Code Interpreter (BETA)\n\nAllows GPT to execute Python code locally. The model has been instructed to strictly only produce safe code,\nbut use at your own risk!\n\nOnly models with function calling capabilities support this feature. Currently, this includes only selected DeepInfra\nmodels.\n\n---\n\n## FAQs\n\n#### Why is this extension not on the official Raycast store?\n\n- By default, this extension comes with a number of third-party providers which you can use for free. Raycast doesn't\n like that because it's not possible to entirely verify the quality of these providers. (I do provide all the privacy\n details [here](Privacy.md), though, and the extension is open-source.) ...and also because they're selling their own\n AI.\n- Also, for me as a developer, publishing updates to the Raycast store is too slow and troublesome; I have to submit a\n PR, wait a week for it to be reviewed, and possibly have it rejected. This is not a good experience for me or for\n users.\n- Thus, the extension will have to be installed from source. Regarding this, I apologize as it's indeed more complicated\n than downloading it from the store. I have tried my best to make the installation process quick and streamlined -\n please do provide feedback on whether it was simple enough!\n\n#### How does this extension compare to the paid Raycast AI? Why should I use this instead?\n\nI’ve noticed many users subscribing to Raycast AI, which is quite expensive, and then only using it for a few casual\nchats a day. Honestly, that’s unnecessary.\n**Here’s my honest suggestion: Everyone should first try out raycast-g4f.** Here’s why:\n\n- **Freedom**. You get to choose whatever provider/API to use, and can even add multiple at once. OpenAI, Anthropic,\n Google…\n basically any API out there. (“Can I use my OpenAI API Key in Raycast AI?” No, you can’t.) That also means your AI\n access isn’t just locked inside Raycast. (“Can I use my Raycast AI in other apps?” No, you can’t.)\n- **Price**. This extension is completely free to use if you're sticking with the built-in providers. And if you want to\n use\n your own API, you pay only for what you use. That’s cheaper than raycast AI. There is no fixed price - you can choose\n whatever provider you want. You can even run models locally and connect them to the extension.\n- **Privacy**. The extension is open source and it respects your privacy. Everything is stored only on your device, and\n you\n choose what data you send.\n\n#### Why is the UI worse than Raycast AI?\n\n- Because the developer API that Raycast provides is limited, it's not possible to replicate the Raycast AI interface\n exactly.\n The components I can use are very simple, and a lot of the features in built-in Raycast commands are not available to\n extension developers.\n- However, I've spent a lot of effort trying to make the UI really intuitive, and I'm always open to feedback on how to\n improve it!\n\n#### The extension isn't working! (e.g. I sent a message but there is no response)\n\n- Sometimes third party providers can be slow or unresponsive. If you're experiencing this issue, please try again in a\n few minutes; or if the problem persists, please try switching to another provider.\n- If you've tried various providers and the issue still persists, please open an issue on GitHub!\n\n#### How do I contribute to the extension?\n\n- I welcome all contributions! If you have an idea for a new feature, or if you've found a bug, please open an issue.\n- If you'd like to contribute code, please open a pull request, and I'll make sure to review it as soon as possible.\n\n---\n\n## Star History\n\n[![Star History Chart](https://api.star-history.com/svg?repos=XInTheDark/raycast-g4f\u0026type=Date)](https://star-history.com/#XInTheDark/raycast-g4f\u0026Date)\n\n## License \u0026 Acknowledgements\n\nLicense: GPLv3. Full license is found in LICENSE.txt.\n\nThe code base is derived from [Raycast Gemini](\nhttps://github.com/raycast/extensions/tree/main/extensions/raycast-gemini) by Evan Zhou.\n\nThird-party libraries used for generation:\n\n- [g4f-image](https://www.npmjs.com/package/g4f-image)\n- [gemini-ai-sdk](https://www.npmjs.com/package/gemini-ai-sdk)\n\n*(Both packages are maintained by the extension author.)*\n\nSome of the code in this repository was inspired or ported from the\noriginal [gpt4free](https://github.com/xtekky/gpt4free) project (written in Python).\n\n\n[active-badge]: https://img.shields.io/badge/Active-brightgreen\n\n[inactive-badge]: https://img.shields.io/badge/Inactive-red\n\n[unknown-badge]: https://img.shields.io/badge/Unknown-grey\n\n[local-api-help]: https://github.com/XInTheDark/raycast-g4f/wiki/Help-page:-Using-Custom-APIs\n\n[discord-invite]: https://discord.gg/u3xTX79DkJ","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzack4dev%2Fllm--raycast_g4f","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fzack4dev%2Fllm--raycast_g4f","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fzack4dev%2Fllm--raycast_g4f/lists"}