arcee-ai/Caller

TEXT GENERATIONConcurrency Cost:2Model Size:32.8BQuant:FP8Ctx Length:32kPublished:Jun 9, 2025License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

Caller (32B) is a 32.8 billion parameter model developed by arcee-ai, based on the Qwen2.5-32B architecture, with a context length of 131072 tokens. It is specifically engineered for managing complex tool-based interactions and API function calls. This model excels at precise execution and intelligent orchestration for sophisticated automation pipelines.

Loading preview...

Caller (32B) Overview

Caller (32B) is a 32.8 billion parameter language model built upon the Qwen2.5-32B architecture. Developed by arcee-ai, this model is specifically designed for robust integrations and optimized for handling intricate tool-based interactions and API function calls. Its core strength lies in its ability to precisely execute tasks, intelligently orchestrate workflows, and facilitate effective communication between diverse systems.

Key Capabilities

  • Tool and API Orchestration: Manages complex interactions with external tools and API function calls.
  • Integration Management: Facilitates seamless integrations between enterprise systems like CRMs and ERPs.
  • Multi-step Workflow Automation: Supports the execution of multi-step workflows, including intelligent condition handling.
  • Real-time Monitoring: Capable of real-time monitoring and diagnostics in various environments, such as IoT or SaaS.

Good For

  • Automating complex business processes requiring interaction with multiple external services.
  • Developing intelligent agents that need to perform actions via API calls.
  • Building sophisticated automation pipelines where precise execution and intelligent decision-making are critical.

Caller (32B) is released under the Apache-2.0 License, allowing for broad use and distribution.