Dozens of Tools. Boundless Creativity.

Generation

A placeholder for an LLM completion.

Structured Generation

Define the output format for an LLM.

Actions

Connect wordware with 2k+ integrations.

Web search with Exa

Query high quality information from the internet.

Website similarity search

Find similar websites to a given url.

Webscrape

Scrape the data from a url.

Image generation

Generate an image using Flux or Stable Diffusion 3.

Text-to-speech with ElevenLabs

Convert text to speech.

Speech-to-text with Deepgram

Convert speech to text.

File Content Extractor

Inject file content into the prompt.

URL Converter

Convert content from a URL into a File, Audio, or Image.

Loop

Repeat a section multiple times.

Code execution

Runs some custom JavaScript or Python code.

If-Else

Add some branching logic.

Flow

Run a linked flow. Previously referred to as a prompt.

Date

Get the current date.

Ask a Human

Runtime input during prompt execution.

Input

Value entered when the program is run.

Image

Upload an image from your computer.

Audio

Upload an audio file from your computer.

PDF

Upload a PDF from your computer.

Send Feedback

Let us know how we can improve.

Text

Just start typing with plain text.

Heading 1

Big section heading.

Heading 2

Medium section heading.

Heading 3

Small section heading.

Comment

A description, not included in the prompt.

Bullet List

Create a simple bullet list.

Numbered List

Create a list with numbering.

Code view

Display a code snippet nicely.

Quote

Capture a quote.

Claude 2

Claude 2.1 is a powerful older generation model from Anthropic with 200k token context capable of complex reasoning, not recommended for new projects

Claude 3 Haiku

A cheap and fast model with 200k token context and vision capabilities

Claude 3 Opus

A top-performing model for reasoning, fluency and understanding with 200k context and vision capabilities

Claude 3 Sonnet

A powerful model and reasonably fast model with 200k token context and vision capabilities

Claude 3.5 Haiku

An upgraded cheap and fast model with 200k token context

Claude 3.5 Sonnet v1

Anthropic's previous most powerful model with 200k token context and vision capabilities

Claude 3.5 Sonnet v2

A powerful model from Anthropic with 200k token context and vision capabilities

Claude 3.7 Sonnet

Anthropic's most powerful model with 200k token context and vision capabilities

Claude 3.7 Sonnet Thinking

Anthropic's most powerful model with extended thinking, 200k token context, and vision capabilities

Claude 3.5 Haiku

Anthropic's most powerful model with 200k token context and vision capabilities

DeepSeek V3

DeepSeek's chat model

DeepSeek R1

DeepSeek's reasoning model

DeepSeek R1 Fast

DeepSeek's reasoning model now even faster with Groq

Command A

A high-performance, enterprise-focused generative model with long context, multilingual support, and efficient deployment on minimal hardware

Command R

A mid-weight model with a long context and the optional online capabilities

Command R Online

A mid-weight model with a long context and online capabilities

Command R+

A powerful model with a 128k context window optimised for RAG and optional online capabilities

Command R+ Online

A powerful model with a 128k context window optimised for RAG and online capabilities

Gemini 2.0 Pro

A very fast multimodal model from Google with a very large 1 million token context. Still experimental, NOT FOR PRODUCTION USE

Gemini 2.0 Pro Online

A very fast multimodal model from Google with a very large 1 million token context. This model will search Google before answering, allowing it to provide responses grounded on up to date information from the internet. Still experimental, NOT FOR PRODUCTION USE

Gemini 2.0 Flash Lite

A very fast multimodal model from Google with a very large 1 million token context.

Gemini 2.0 Flash

A very fast multimodal model from Google with a very large 1 million token context.

Gemini 2.0 Flash Online

A very fast multimodal model from Google with a very large 1 million token context. This model will search Google before answering, allowing it to provide responses grounded on up to date information from the internet.

Gemini 1.5 Flash

A fast and powerful multimodal model from Google with a huge 1 million token context

Gemini 1.5 Flash online

A fast and powerful multimodal model from Google with a huge 1 million token context. This model will search Google before answering, allowing it to provide responses grounded on up to date information from the internet.

Gemini 1.5 Pro

Google's most powerful model with a huge 1 million token context as well as vision and audio capabilities

Gemini 1.5 Pro online

Google's most powerful model with a huge 1 million token context as well as vision and audio capabilities. This model will search Google before answering, allowing it to provide responses grounded on up to date information from the internet.

Gemini Pro

Gemini 1.0 Pro model from Google with vision and audio capabilities

Gemini Pro

Gemini 1.0 Pro model from Google with vision and audio capabilities

Gemma 2B

A super fast and capable small model from Google

Gemma 2 9B

A fast and capable small model from Google

Gemma 2 27B

A larger, more capable model from Google

GPT-3.5

The GPT-3.5 model used in the free version of ChatGPT - fast and effective for most needs

GPT-4

A more powerful but slower model - useful when complex reasoning is required

GPT-4o

As smart as GPT-4 Turbo but faster and half the cost

GPT-4o mini

New intelligent and affordable model from OpenAI, with vision capabilities

GPT-4 Turbo

A powerful model with 128k token context, ~3x cheaper than GPT-4 with speed similar to GPT-3.5 and vision capabilities

GPT-4 Vision

A powerful model with 128k token context, ~3x cheaper than GPT-4 with speed similar to GPT-3.5 and vision capabilities

OpenAI o1 Mini

Preview: Faster and cheaper reasoning model particularly good at coding, math and science. Note that this model is in preview, and should not be used in production. There are likely bugs and usage restrictions.

o1-preview

Preview: Reasoning model designed to solve hard problems across domains. Note that this model is in preview, and should not be used in production. There are likely bugs and usage restrictions.

o1

Reasoning model with vision capabilities

o3 Mini

OpenAI's newest reasoning model, featuring faster and cheaper reasoning

GPT-4.5 Preview

A research preview of GPT-4.5, OpenAI's newest and best model for chat yet. Still experimental, NOT FOR PRODUCTION USE

GPT-4o Search

A research preview of GPT-4o Search, OpenAI's newest and best model for search yet. Still experimental, NOT FOR PRODUCTION USE

GPT-4o Mini Search

A research preview of GPT-4o Mini Search, OpenAI's cheapest and fastest model for search yet. Still experimental, NOT FOR PRODUCTION USE

ChatGPT-4o

ChatGPT-4o is a fine tuned version of GPT-4o with a focus on chat and conversation. Used in ChatGPT.

Llama 3 70B

A state of the art open model from Meta

Llama 3 70B fast

A state of the art open model from Meta with groq speed

Llama 3 8B

A small, fast and powerful open model from Meta

Llama 3 8B fast

A small, fast and powerful open model from Meta with groq speed

Llama 3.1 405B

The largest model from Meta, and one of the most powerful available

Llama 3.1 70B

Highly performant, cost effective model from Meta

Llama 3.1 70B fast

Highly performant, cost effective model from Meta, now with groq speed

Llama 3.1 8B

Light-weight, fast model from Meta

Llama 3.1 8B Fast

Light-weight, fast model from Meta, now even faster with groq

Llama 3.2 3B

Super light-weight, fast model from Meta

Llama 3.2 11B

A light-weight, fast model from Meta

Llama 3.2 90B

A powerful vision enabled model from Meta

Llama 3.3 70B

Highly performant, text-only, cost effective model from Meta

Llama 3.3 70B Fast

Highly performant, text-only, cost effective model from Meta, now even faster with groq

Mistral Medium

A powerful model from Mistral AI

Mistral Large 2

Mistral AI's most powerful model, and one of the best models available

Mistral NeMo

Mistral AI's best small model. A state-of-the-art 12B model with 128k context length

Mixtral 8x22B

An 8x22B mixture of expert model from Mistral AI

Mixtral 8x7B fast

An 8x7B mixture of expert model from Mistral AI. Excellent performance with super fast inference through groq

Mistral Small 3

A latency-optimized model from Mistral AI with 24B parameters that is fast and cheap.

Mistral 7B

A 7B parameter model, fast inference and ~100x cheaper than GPT-4 turbo. This specific model is Mistral 7B Instruct

Mixtral 8x7B

An 8x7B mixture of expert model from Mistral AI. Excellent performance with fast inference

Nova Pro

Amazons largest multimodal model of the Nova family yet

Nova Lite

A low-cost multimodal model by Amazon

Nova Micro

A small and fast text-only model by Amazon

OpenChat

A cheap and fast 7B open source parameter model that outperforms OpenAI's GPT-3.5 on many metrics

Sonar Huge

The most reasoning-capable model from Perplexity AI with up-to-date information. Based on Llama 3.1 405B

Sonar Large

A model from Perplexity AI with up-to-date information based on Llama 3.1 70B

Sonar Small

A model from Perplexity AI with up-to-date information based on Llama 3.1 8B

Sonar

A lightweight model from Perplexity AI with up-to-date information. Performs one search per query.

Sonar Pro

A powerful model from Perplexity AI with up-to-date information

Sonar Reasoning

A powerful reasoning model from Perplexity AI with up-to-date information

Sonar Reasoning Pro

Perplexity's premier reasoning offering powered by DeepSeek R1 with Chain of Thought (CoT).

Sonar Deep Research

Perplexity's deep research model conducts comprehensive, expert-level research and synthesizes it into accessible, actionable reports. Still experimental, NOT FOR PRODUCTION USE

R1-1776

A version of the DeepSeek R1 model that has been post-trained to provide uncensored, unbiased, and factual information.

DBRX

A mixture-of-experts model trained by Databricks

Reka Core

Superior capabilities for complex tasks

Reka Flash

Fast and cost-efficient multi-modal model for most tasks

Grok 2

The flagship model from xAI

Grok 2 Vision

The vision-capable model from xAI. Still experimental, NOT FOR PRODUCTION USE