{"id":15057217,"url":"https://github.com/bytez-com/docs","last_synced_at":"2025-04-10T05:08:39.919Z","repository":{"id":243141684,"uuid":"806804556","full_name":"Bytez-com/docs","owner":"Bytez-com","description":"Access 40,000+ open and closed-source AI models across 30+ tasks with a unified schema. Run fast, affordable inference in seconds through a simple API ✨","archived":false,"fork":false,"pushed_at":"2025-03-31T21:52:59.000Z","size":7604,"stargazers_count":9,"open_issues_count":2,"forks_count":1,"subscribers_count":5,"default_branch":"main","last_synced_at":"2025-04-10T05:08:33.436Z","etag":null,"topics":["ai","api","bytez","deepseek","falcon","javascript","julia","llama","llms","mistral","multimodla","os","phi3","python"],"latest_commit_sha":null,"homepage":"https://bytez.com/","language":"TypeScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Bytez-com.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"2024-05-28T00:09:03.000Z","updated_at":"2025-04-02T05:32:03.000Z","dependencies_parsed_at":"2024-06-18T23:18:33.435Z","dependency_job_id":"db775d92-f42a-4e0d-9615-080ed6c78227","html_url":"https://github.com/Bytez-com/docs","commit_stats":null,"previous_names":["bytez-com/docs"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Bytez-com%2Fdocs","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Bytez-com%2Fdocs/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Bytez-com%2Fdocs/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Bytez-com%2Fdocs/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Bytez-com","download_url":"https://codeload.github.com/Bytez-com/docs/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248161271,"owners_count":21057555,"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","bytez","deepseek","falcon","javascript","julia","llama","llms","mistral","multimodla","os","phi3","python"],"created_at":"2024-09-24T22:03:56.633Z","updated_at":"2025-04-10T05:08:39.897Z","avatar_url":"https://github.com/Bytez-com.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n \u003cimg alt=\"bytez\" height=\"250px\" src=\"https://github.com/Bytez-com/docs/assets/9612780/610ae3a1-65b5-4f8a-8ed5-0bae9134ab5f\"\u003e\n\n[![](https://dcbadge.limes.pink/api/server/https://discord.gg/Zrd5UbMEBA)](https://discord.com/invite/Z723PfCFWf) \n[![](https://img.shields.io/badge/Bytez-000000?style=for-the-badge\u0026logo=x\u0026=logoColor=white)](https://x.com/Bytez)\n[![Discord](https://img.shields.io/discord/844731722700816395)](https://discord.com/invite/Z723PfCFWf) \n[![NPM Version](https://img.shields.io/npm/v/bytez.js)](https://www.npmjs.com/package/bytez.js)\n[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1oZ4_yQoryL9a3CCLiY29JpEI1L5uwqO-?authuser=1#scrollTo=3LRTz2egUNh7\u0026uniqifier=3)\n\n\u003c/div\u003e\n\n# Bytez Model API\n\n✨ Access 40k+ open-source and proprietary AI models through a standard API. Achieve GPU-backed performance at CPU pricing ✨\n\n## Table of Contents\n\n1. [Introduction](#introduction)\n2. [Quickstart](#quickstart)\n3. [API Documentation](#api-documentation)\n   - [Get a Bytez API Key](#get-a-bytez-api-key)\n   - [List Models \u0026 Tasks](#list-models--tasks)\n   - [Play with Trending Models on Bytez](#play-with-trending-models-on-bytez)\n4. [Libraries](#libraries)\n5. [Docker](#docker)\n6. [Capabilities](#capabilities)\n   - [Chat (Text Generation)](#chat)\n   - [Chat + Vision](#chat--vision)\n   - [Chat + Video](#chat--video)\n   - [Chat + Audio](#chat--audio)\n   - [Image Generation](#image-generation)\n   - [Embeddings](#embeddings)\n   - [Function Calling](#function-calling)\n   - [Streaming](#streaming)\n   - [All Model Tasks](#all-model-tasks)\n7. [Proprietary Models](#proprietary-models)\n8. [Pricing](#pricing)\n9. [API Playground](#api-playground)\n10. [Status](#status)\n11. [Resources](#resources)\n12. [Feedback](#feedback)\n\n---\n\n## Introduction\n\nBytez Model API streamlines integration with 40k+ open-source and proprietary AI models across 33 ML tasks. By standardizing inputs for `text`, `images`, `audio`, and more, it eliminates the complexity of inconsistent formats, enabling developers to effortlessly interact with models for tasks like `chat`, `text generation`, `image generation`, `video generation`, and beyond.\n\n---\n\n## Quickstart\nGet your API Key by signing up on [Bytez](http://bytez.com), then navigating to Settings in your account.\n\nValidate by running an inference:\n\n#### REST\n```bash\ncurl --location 'https://api.bytez.com/models/v2/openai-community/gpt2' \\\n--header 'Authorization: Key BYTEZ_API_KEY' \\\n--header 'Content-Type: application/json' \\\n--data '{\n    \"text\": \"Dreams are messages from the \"\n}'\n```\n\n## API Documentation\n\n### Get a Bytez API Key\n\n1. Log into [Bytez](http://bytez.com).\n2. Navigate to the `Settings` page.\n3. Locate your API key under the **API Keys** section and copy it.\n\nUse this key in the `Authorization` header for all API requests:\n\n```http\nAuthorization: Key your-key-here\n```\n\n### Send Your First API Request\nYou can use a `curl` command to verify your setup:\n\n```bash\ncurl --location 'https://api.bytez.com/models/v2/NousResearch/Hermes-3-Llama-3.1-8B' \\\n--header 'Authorization: Key BYTEZ_API_KEY' \\\n--header 'Content-Type: application/json' \\\n--data '{\n    \"messages\": [\n        {\n            \"role\": \"system\",\n            \"content\": \"You'\\''re a helpful assistant\"\n        },\n        {\n            \"role\": \"user\",\n            \"content\": \"Dreams are messages from the \"\n        }\n    ]\n}'\n```\n\nSomething not right or need another API Key? DM our team in [Discord](https://discord.com/invite/Z723PfCFWf) and we'll resolve.\n\n### Accessing Closed Source Models\nYou can interact with proprietary [chat](chat/text#proprietary-models) models by `OpenAI`, `Anthropic`, `Cohere`, `Google`, and `Mistral`\n\nTo use these models, you'll need two keys:\n\n1. Your `Bytez API Key`: Obtained as described above.\n2. `Provider Key`: The key specific to the provider you want to access (e.g., OpenAI API key).\n\nExample Headers\n```http\nAuthorization: Key your-bytez-api-key\nProvider-key: your-provider-key\n```\n\n### Notes\n1. `No Additional Charges`: Bytez does not charge for accessing proprietary models; however, the respective provider's billing applies.\n2. `Seamless Integration`: You can interact with closed-source models using the same standardized input structure as open-source models.\n\n\n### List Models \u0026 Tasks\n#### Python\n```python\nfrom bytez import Bytez\n\nclient = Bytez(\"YOUR_BYTEZ_KEY_HERE\")\n\n## List all models\nmodels = client.list_models()\nprintln(models)\n\n## List models by task\nmodels_by_task = client.list_models(\"object-detection\")\nprintln(models_by_task)\n```\n#### Javascript\n```javascript\nimport Bytez from \"bytez.js\";\n\nconst client = new Bytez(\"YOUR_BYTEZ_KEY_HERE\");\n// List all models\nconst models = await client.list.models.all();\nconsole.log(models);\n// List models by task\nconst modelsByTask = await client.list.models.byTask(\"object-detection\");\n\nconsole.log(modelsByTask);\n```\n#### Julia\n```julia\nusing Bytez\nclient = Bytez(\"YOUR_BYTEZ_KEY_HERE\")\nmodel_list = client.list_models()\n\nprintln(model_list)\n```\n#### REST\n```bash\ncurl --location 'https://api.bytez.com/models/v2/list/models' \\\n--header 'Authorization: Key YOUR_BYTEZ_KEY_HERE'\n\ncurl --location 'https://api.bytez.com/models/v2/list/models?task=chat' \\\n--header 'Authorization: Key YOUR_BYTEZ_KEY_HERE'\n```\n### Play with Trending Models on Bytez\nWe have an [API playground](https://docs.bytez.com/model-api/playground/overview) to demo over 40k models across 33 tasks. Or, feel free to play with models on the [Bytez platform](https://bytez.com/).\n\n## Libraries\nUsing `Python 3.9+`, `JavaScript`, or `Julia`, install the appropriate package:\n\n### Python\n```python \npip install bytez\n```\n\n### Javascript\n```javascript \nnpm install bytez.js\n```\n### Julia\n```julia julia\n// Run the command julia\n// Press ]\n// Run the command below\nadd Bytez\n```\n\n### Run Inference\n\n#### Python\n```python\nfrom bytez import Bytez\n\nclient = Bytez(\"YOUR_BYTEZ_KEY_HERE\")\nmodel = client.model(\"Qwen/Qwen2-7B-Instruct\")\nmodel.load()\n\ninput_text = \"Dreams are messages from the \"\nmodel_params = {\"max_new_tokens\": 20, \"max_new_tokens\": 5, \"temperature\": 0.5}\n\nresult = model.run(input_text, model_params=model_params)\noutput = result[\"output\"]\ngenerated_text = output[0][\"generated_text\"]\nprint(generated_text)\n```\n\n#### JavaScript\n```javascript\nimport Bytez from \"bytez.js\";\n\nconst client = new Bytez(\"YOUR_BYTEZ_KEY_HERE\");\nconst model_id = \"openai-community/gpt2\";\nconst model = client.model(\"openai-community/gpt2\");\n\nawait model.load();\n\nconst output = await model.run(\"Dreams are messages from the \", {\n  max_new_tokens: 20,\n  min_new_tokens: 5\n});\n\nconsole.log(output);\n```\n#### Julia\n```julia\nusing Bytez\n\nclient = Bytez(\"YOUR_BYTEZ_KEY_HERE\")\n\nmodel = client.model(\"Qwen/Qwen2-7B-Instruct\")\n\nmodel.load()\n\ninput_text = \"Dreams are messages from the \"\n\noptions = Dict(\n\t\"params\" =\u003e Dict(\n\t\t\"max_new_tokens\" =\u003e 20,\n\t\t\"min_new_tokens\" =\u003e 5,\n\t\t\"temperature\" =\u003e 0.5,\n\t)\n)\n\nresult = model.run(input_text, options)\noutput = result[\"output\"]\ngenerated_text = output[1][\"generated_text\"]\nprintln(generated_text)\n```\n## Docker\nAll Bytez model images are available on [Docker Hub](https://hub.docker.com/u/bytez), models can be played with via our [Models](https://bytez.com/models) page 🤙\n\n### Image Source Code\nThe source code that runs for a given model in the docker image can be found [here](https://github.com/Bytez-com/models)\n\n## Capabilities \n### Chat\nGenerate text with chat models using structured inputs.\n\n#### Javascript\n```javascript javascript\n\nimport Bytez from \"bytez.js\";\nconst client = new Bytez(\"YOUR_BYTEZ_KEY_HERE\");\n\nconst messages = [\n  {\n    role: \"system\",\n    content: \"You are a friendly chatbot\",\n  },\n  {\n    role: \"user\",\n    content: \"What is the capital of England?\",\n  },\n];\n\nconst model = client.model(\"microsoft/Phi-3-mini-4k-instruct\");\n\nawait model.load();\n\nconst { output } = await model.run(messages, { max_length: 100 });\nconst [{ generated_text }] = output;\n\nfor (const message of generated_text) {\n  console.log(message);\n  const { content, role } = message;\n  console.log({ content, role });\n}\n```\nFull documentation [here](https://docs.bytez.com/model-api/docs/chat/text)\n\n\n### Chat + Vision\nUse chat models with images as input to generate text-based responses.\n\n#### Javascript\n```javascript javascript\nconst Bytez = require(\"bytez.js\");\nconst client = new Bytez(\"YOUR_BYTEZ_KEY_HERE\");\n\nconst model = client.model(\"meta-llama/Llama-3.2-11B-Vision-Instruct\");\nawait model.load();\n\nconst textInput = [\n  {\n    role: \"system\",\n    content: [{ type: \"text\", text: \"You are a helpful assistant.\" }]\n  },\n  {\n    role: \"user\",\n    content: [\n      { type: \"text\", text: \"What is this image?\" },\n      { type: \"image\", url: \"https://hips.hearstapps.com/hmg-prod/images/how-to-keep-ducks-call-ducks-1615457181.jpg?crop=0.670xw:1.00xh;0.157xw,0\u0026resize=980:*\" }\n    ]\n  }\n];\n\nconst { output } = await model.run(textInput);\nconsole.log(output);\n```\n\nFull documentation [here](https://docs.bytez.com/model-api/docs/chat/vision)\n\n### Chat + Video\nUse chat models with video input to generate insightful responses.\n\n#### Javascript\n```javascript javascript\nconst Bytez = require(\"bytez.js\");\nconst client = new Bytez(\"YOUR_BYTEZ_KEY_HERE\");\n\nconst model = client.model(\"llava-hf/LLaVA-NeXT-Video-7B-hf\");\nawait model.load();\n\nconst textInput = [\n  {\n    role: \"system\",\n    content: [{ type: \"text\", text: \"You are a helpful assistant.\" }]\n  },\n  {\n    role: \"user\",\n    content: [\n      { type: \"text\", text: \"Why is this video funny?\" },\n      { type: \"video\", url: \"https://example.com/path-to-video.mp4\" }\n    ]\n  }\n];\n\nconst { output } = await model.run(textInput);\nconsole.log(output);\n\n```\n\nFull documentation [here](https://docs.bytez.com/model-api/docs/chat/video)\n\n### Chat + Audio\nProcess and analyze audio inputs with chat models.\n\n#### Javascript\n```javascript javascript\nconst Bytez = require(\"bytez.js\");\nconst client = new Bytez(\"YOUR_BYTEZ_KEY_HERE\");\n\nconst model = client.model(\"Qwen/Qwen2-Audio-7B-Instruct\");\nawait model.load();\n\nconst textInput = [\n  {\n    role: \"system\",\n    content: [{ type: \"text\", text: \"You are a helpful assistant.\" }]\n  },\n  {\n    role: \"user\",\n    content: [\n      { type: \"text\", text: \"What sound is this?\" },\n      { type: \"audio\", url: \"https://example.com/path-to-audio.mp3\" }\n    ]\n  }\n];\n\nconst { output } = await model.run(textInput);\nconsole.log(output);\n```\nFull documentation [here](https://docs.bytez.com/model-api/docs/chat/audio)\n\n### Image Generation\nGenerate images using Bytez API with `base64` or `URL` inputs.\n\n#### Javascript\n```javascript javascript\nimport Bytez from \"bytez.js\";\nimport { dirname } from \"path\";\nimport { fileURLToPath } from \"url\";\nimport { writeFileSync } from \"node:fs\";\n\nconst __filename = fileURLToPath(import.meta.url);\nconst __dirname = dirname(__filename);\n\nconst client = new Bytez(\"YOUR_BYTEZ_KEY_HERE\");\n\nconst model = client.model(\"dreamlike-art/dreamlike-photoreal-2.0\");\n\nawait model.load();\n\nconst { output_png } = await model.run(\n  \"A beautiful landscape with mountains and a river\"\n);\n\nconst buffer = Buffer.from(output_png, \"base64\");\n\n// Write the image to the local file system\nwriteFileSync(`${__dirname}/output.png`, buffer);\n```\n\n### Embeddings\nGenerate `text` and `vector` embeddings\n\n```javascript javascript\nimport Bytez from \"bytez.js\";\n\nconst client = new Bytez(\"API_KEY\");\n\n// 1) Select the model\nconst model = client.model(\"nomic-ai/nomic-embed-text-v1.5\");\n\n// 2) Load the model\nawait model.load();\n\n// 3) Run the model\nconst output = await model.run(\"Once upon a time\");\n\nconsole.log(output);\n\n```\nFull documentation [here](https://docs.bytez.com/model-api/docs/embeddings)\n\n### Function Calling\nExecute `code` or `actions` based on model-generated outputs\n\n```javascript javascript\nimport Bytez from \"bytez.js\";\n\nconst client = new Bytez(\"YOUR_BYTEZ_KEY_HERE\");\n\nconst inputText = \"What's the weather like in Seattle right now?\";\n\nconst modelParams = {\n  max_new_tokens: 2000,\n  min_new_tokens: 50,\n  temperature: 0.001,\n  do_sample: false\n};\n\nconst promptTemplate = `\nFunction:\ndef get_weather_data(coordinates):\n    \"\"\"\n    Fetches weather data from the Open-Meteo API for the given latitude and longitude.\n\n    Args:\n    coordinates (tuple): The latitude and longitude of the location.\n\n    Returns:\n    float: The current temperature in the coordinates you've asked for\n    \"\"\"\n\nFunction:\ndef get_coordinates_from_city(city_name):\n    \"\"\"\n    Fetches the latitude and longitude of a given city name using the Maps.co Geocoding API.\n\n    Args:\n    city_name (str): The name of the city.\n\n    Returns:\n    tuple: The latitude and longitude of the city.\n    \"\"\"\n\nUser Query: {query}\u003chuman_end\u003e\n`;\n\nconst model = client.model(\"Nexusflow/NexusRaven-V2-13B\");\n\nawait model.load();\n\nconst prompt = promptTemplate.replace(\"{query}\", inputText);\n\nconst stream = await model.run(prompt, { stream: true, params: modelParams });\n\nconst textStream = stream.pipeThrough(new TextDecoderStream());\n\nfor await (const chunk of textStream) {\n  console.log(chunk);\n}\n\n```\nFull documentation [here](https://docs.bytez.com/model-api/docs/function-calling)\n\n## Streaming\n\n\u003cTip\u003eStreaming allows you to receive model outputs incrementally as soon as they are available, which is ideal for tasks like real-time responses or large outputs.\u003c/Tip\u003e\n\n### How Streaming Works\n\nTo enable streaming, pass `true` as the third argument to the `model.run()` function. The model will return a stream that you can read incrementally.\n\n```javascript javascript\nconst stream = await model.run(textInput, params, true);\n```\n\n### Node.js \n\n```javascript javascript\nconst { Readable } = require('stream');\n\nconst stream = await model.run(textInput, params, true);\n\ntry {\n  const readableStream = Readable.fromWeb(stream); // Convert Web Stream to Node.js Readable Stream\n  for await (const chunk of readableStream) {\n    console.log(chunk.toString()); // Handle each chunk of data\n  }\n} catch (error) {\n  console.error(error); // Handle errors\n}\n```\n### Browser \n```javascript javascript\nconst stream = await model.run(textInput, params, true);\n\ntry {\n  const reader = stream.getReader(); // Get a reader for the Web Stream\n\n  while (true) {\n    const { done, value } = await reader.read(); // Read the stream chunk-by-chunk\n    if (done) break; // Exit when the stream ends\n    console.log(new TextDecoder().decode(value)); // Convert Uint8Array to string\n  }\n} catch (error) {\n  console.error(error); // Handle errors\n}\n```\n\n## All Model Tasks\nOur API provides access to a wide range of pretrained models across 33 machine learning tasks, each tailored to specific applications like `summarization`, `document question-answering`, `audio classification`, and more. \n\nExplore the full list of tasks [here](http://docs.bytez.com/model-api/docs/all-tasks/audio-classification).\n\n## Proprietary Models\nOur v2 endpoint supports interacting with proprietary models from `Anthropic`, `Google`, `Cohere`, `OpenAI`, and `Mistral`.\n\n### OpenAI\n```bash\ncurl --location 'https://api.bytez.com/models/v2/openai/gpt-4o-mini' \\\n--header 'Authorization: Key YOUR_BYTEZ_KEY_HERE' \\\n--header 'Provider-Key: PROVIDER_KEY' \\\n--header 'Content-Type: application/json' \\\n--data '{\n    \"messages\": [{\"role\": \"user\", \"content\": \"Hello my name is Bob and I like to eat\"}],\n    \"stream\": false,\n    \"params\": { \"max_tokens\": 100 }\n}'\n```\n### Google Gemini\n```bash\ncurl --location 'https://api.bytez.com/models/v2/google/gemini-1.5-flash' \\\n--header 'Authorization: Key YOUR_BYTEZ_KEY_HERE' \\\n--header 'Provider-Key: PROVIDER_KEY' \\\n--header 'Content-Type: application/json' \\\n--data '{\n    \"messages\": [{\"role\": \"user\", \"content\": \"Hello my name is Bob and I like to eat\"}],\n    \"stream\": false,\n    \"params\": { \"temperature\": 1 }\n}'\n```\n### Cohere\n```bash\ncurl --location 'https://api.bytez.com/models/v2/cohere/command-r' \\\n--header 'Authorization: Key YOUR_BYTEZ_KEY_HERE' \\\n--header 'Provider-Key: PROVIDER_KEY' \\\n--header 'Content-Type: application/json' \\\n--data '{\n    \"messages\": [{\"role\": \"user\", \"content\": \"Cats and rabbits who reside in fancy little houses\"}],\n    \"stream\": false,\n    \"params\": { \"max_tokens\": 50 }\n}'\n```\n### Mistral\n```bash \ncurl --location 'https://api.bytez.com/models/v2/mistral/mistral-small-latest' \\\n--header 'Authorization: Key YOUR_BYTEZ_KEY_HERE' \\\n--header 'Provider-Key: PROVIDER_KEY' \\\n--header 'Content-Type: application/json' \\\n--data '{\n    \"messages\": [{\"role\": \"user\", \"content\": \"Cats and rabbits who reside in fancy little houses\"}],\n    \"stream\": false,\n    \"params\": { \"max_tokens\": 50 }\n}'\n```\n### Anthropic\n```bash \ncurl --location 'https://api.bytez.com/models/v2/anthropic/claude-3-haiku-20240307' \\\n--header 'Authorization: Key YOUR_BYTEZ_KEY_HERE' \\\n--header 'Provider-Key: PROVIDER_KEY' \\\n--header 'Content-Type: application/json' \\\n--data '{\n    \"messages\": [{\"role\": \"user\", \"content\": \"Cats and rabbits who reside in fancy little houses\"}],\n    \"stream\": false,\n    \"params\": { \"max_tokens\": 50 }\n}'\n```\n## Pricing\nInference pricing for models is designed to be straightforward and predictable. Instead of relying on complex token-based pricing (which doesn't make sense for non-text-generation models), we calculate costs based on `Inference Meter Price` and `Time to First Inference`.\u003c/Info\u003e\n\n### Formula\n\n```plaintext\nPricing = Meter Price × Inference Time\n```\n\n### Key Features\n\n### **Instance-Based Pricing**\n- Models run on **instances** optimized for **RAM usage**.\n- Instances are categorized by size (e.g., `Micro`, `Small`, `Super`).\n- **LLMs (Large Language Models)** have their own specific pricing meters.\n\n### **Transparent API Response Metadata**\nEach API response includes:\n- **`Inference Meter`**\n- **`Inference Meter Price`**\n- **`Inference Time`**\n- **`Inference Cost`**\n\n### Prices\n\n### Language Models\n\n| Instance Size | GPU RAM (GB) | Inference Meter Price ($/sec) |\n|------------------|------------------|-------------------------------|\n| Micro         | 16               | 0.0000872083                  |\n| XS            | 24               | 0.0001475035                  | \n| SM            | 64               | 0.0006478333                  | \n| MD            | 96               | 0.0008433876                  | \n| LG            | 128              | 0.0012956667                  | \n| XL            | 192              | 0.0024468774                  | \n| XXL           | 320              | 0.0047912685                  | \n| Super         | 640              | 0.0059890856                  |\n\n\n### All other models\n\n| Instance Size | GPU RAM (GB) | Inference Meter Price ($/sec) | \n|------------------|------------------|-------------------------------|\n| Micro            | 16               | 0.00053440                    |\n| XS               | 24               | 0.00066800                    | \n| SM               | 64               | 0.00427520                    | \n| MD               | 96               | 0.00480960                    | \n| LG               | 128              | 0.00855040                    | \n| XL               | 192              | 0.01603200                    |\n| XXL              | 320              | 0.02458240                    | \n| Super            | 640              | 0.02992640                    | \n\n\n## API Playground\nExplore our API endpoints in the documentation [here](https://docs.bytez.com/model-api/playground/overview).\n\n## Status\nCheck out the status of our [API](https://status.bytez.com)\n\n## Resources\nGet to know our story, our mission, and our roadmap [here](https://docs.bytez.com/company/about).\n\n## Feedback\nWe’re committed to building the best developer experience for AI builders. Have feedback? Let us know on [Discord](https://discord.com/invite/Z723PfCFWf) or open an issue on [GitHub](https://github.com/Bytez-com/docs/issues).\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbytez-com%2Fdocs","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbytez-com%2Fdocs","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbytez-com%2Fdocs/lists"}