Skip to content

GooseAI

GooseAI provides fully managed NLP-as-a-Service via API, offering classic NLP use cases and fast generation speeds at a competitive cost.

Paid

stork
I am a flat drawing.
One flat drawing and one typed sentence.
This drawing was never animated.
A photo, a cartoon, a painting or a 3D character. From 10 cents a minute. Free to try.
Make a photo talkDevelopers: there is an API

Pricing Plans

Small$0.000110 / per-request

GPT-Neo 1.3B · Fairseq 1.3B

Medium$0.000450 / per-request

GPT-J 6B · Fairseq 6B

Large$0.001250 / per-request

Fairseq 13B

Massive$0.002650 / per-request

GPT-NeoX 20B

What is GooseAI?

GooseAI is an NLP-as-a-Service tool developed by CoreWeave and Anlatan (Joint Venture Partners) that enables creative technologists building products on top of large language models and developers looking to reduce AI infrastructure costs to deploy NLP services and serve open-source language models via API. It offers classic NLP use cases and fast generation speeds at a competitive cost. Established in 2022, GooseAI functions as an AI-as-a-Service (AIaaS) platform designed to automate complex workflows and tasks, extending beyond simple code suggestions to execute commands, edit files, and perform multi-step operations. The platform is available as a native desktop application for macOS, Linux, and Windows, a command-line interface (CLI), and an API for embedding. The project has transitioned from block/goose to the Agentic AI Foundation (AAIF) at the Linux Foundation, signifying a move towards broader community governance and development.

Key Features of GooseAI

GooseAI provides a comprehensive set of features designed for developers and creative technologists, focusing on managed NLP services, cost efficiency, and extensibility. Its architecture supports various deployment methods and integrates advanced protocols for enhanced functionality.

  • Fully managed NLP-as-a-Service via API.
  • Support for classic NLP use cases including Text Completion/Generation, Question/Answer, and Classification.
  • Fast generation speeds for AI models, optimizing performance for various tasks.
  • Competitive pricing structure for token generation, offering significant cost savings.
  • API access for seamless integration into existing applications and development workflows.
  • Support for serving open-source language models such as 125M GPT-Neo, 6B GPT-J, and 20B GPT-NeoX.
  • Local execution capabilities for enhanced data privacy and security, operating directly on user machines.
  • Deep integration of the Model Context Protocol (MCP) for tool extensibility, supporting over 1,000 MCP servers.
  • Available as a native desktop application for macOS, Linux, and Windows, alongside a command-line interface (CLI).
  • Automates complex workflows and tasks, including coding, file editing, and multi-step operations.

Who Should Use GooseAI?

GooseAI is designed for a diverse range of users, from individual developers to enterprise teams, who require efficient, cost-effective, and flexible AI solutions for natural language processing and workflow automation.

  • Creative technologists building products on top of large language models, leveraging its API for text generation and NLP services.
  • Developers looking to reduce AI infrastructure costs by utilizing GooseAI's competitive pricing for serving open-source models.
  • Software developers automating repetitive coding tasks, managing complex pipelines, writing and debugging code, and handling deployments.
  • Enterprises and internal teams (e.g., Block) streamlining workflow automation, research, data analysis, and content asset management.
  • Users requiring local execution for sensitive data handling, ensuring enhanced security and control over their AI operations.

How to Use GooseAI

To begin using GooseAI, users can access its API for integration into their applications or download the native desktop client for local operations. The platform provides comprehensive documentation to facilitate setup and model utilization.

  1. Access the GooseAI API documentation at https://goose.ai/docs/api-reference/engines to understand available endpoints and parameters.
  2. Select a desired language model (e.g., 6B GPT-J, 20B GPT-NeoX) and review its specific per-token pricing for output.
  3. Integrate the API into applications using supported programming languages such as Python or JavaScript.
  4. Configure API calls, specifying parameters like the maximum number of tokens generated per API call (up to 2,048 tokens) to manage costs.
  5. For local operations and enhanced privacy, download and install the native desktop app for macOS, Linux, or Windows.
  6. Utilize the Model Context Protocol (MCP) for integrating with over 1,000 available MCP servers to extend tool capabilities, especially for coding agents.

GooseAI Pricing & Plans

GooseAI operates on a paid, usage-based business model, charging based on tokens generated per API call and offering tiered request pricing. The platform's pricing structure is designed to be competitive, aiming for up to 30% cost savings compared to alternatives. While a base price includes the first 25 tokens, costs are primarily determined by output tokens and the selected model.

  • Small: $0.000110 per request
  • Medium: $0.000450 per request
  • Large: $0.001250 per request
  • Massive: $0.002650 per request
  • 125M GPT-Neo, Fairseq: $0.001 per 1k output tokens
  • 1.3B GPT-Neo, Fairseq: $0.003 per 1k output tokens
  • 2.7B GPT-Neo, Fairseq: $0.008 per 1k output tokens
  • 6B GPT-J, Fairseq: $0.012 per 1k output tokens
  • 13B Fairseq: $0.036 per 1k output tokens
  • 20B GPT-NeoX: $0.063 per 1k output tokens

Pros

  • Cost-effective NLP-as-a-Service, offering up to 30% savings compared to alternatives.
  • Supports a range of open-source language models (e.g., GPT-Neo, GPT-J, GPT-NeoX) via API.
  • Provides local execution capabilities for enhanced data privacy and security, operating on user machines.
  • Offers fast generation speeds for various NLP tasks, optimizing performance.
  • Features a modular and extensible design with deep integration of the Model Context Protocol (MCP).
  • Available as a native desktop application (macOS, Linux, Windows) and a command-line interface (CLI).

Cons

  • Primarily focuses on open-source models, potentially lacking the cutting-edge proprietary models offered by some competitors.
  • While cost-effective, it is not free and requires payment for usage based on tokens and requests.
  • For trivial changes or simple tasks, the process of using the AI agent might not always be faster than manual execution.
  • API rate limits are not explicitly detailed beyond token configuration, requiring users to carefully manage costs and usage.
  • The platform's primary focus on NLP-as-a-Service might not cover all advanced multimodal AI capabilities offered by larger cloud providers.

GooseAI vs Competitors

GooseAI positions itself as a cost-effective and flexible alternative in the AI infrastructure market, particularly for NLP-as-a-Service and serving open-source models. Its competitive advantages include local execution capabilities and a focus on reducing AI infrastructure costs.

GooseAI

Paid

From the Stork blog

Enjoying this? Get one like it in your inbox each morning.

one email a day · unsubscribe in two clicks · no third-party tracking