For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
DocsAPI ReferenceSDKs
DocsAPI ReferenceSDKs
  • Introduction
    • Product Overview
    • Pricing
    • Model Explanation
    • Available Models
  • Guides
    • Quickstart
    • Authentication
    • BYOK
    • Dashboard Guide
    • Prompt Templates
    • Embeddings
    • RAG (Files & Search)
    • Audio
    • Images & Vision
    • Image Generation
    • Compare
    • Batch API
    • Auto Routing
    • Realtime Audio
  • Infrastructure
    • Architecture
LogoLogo
On this page
  • Understanding Models
  • Base Models
  • Naming Convention
  • Capabilities & Context Limits
  • Choosing the Right Model
Introduction

Model Explanation

||View as Markdown|
Was this page helpful?
Edit this page
Previous

Pricing Details

Next

Available Models

Built with

Understanding Models

MeshAPI acts as a router, forwarding your standardized API calls to an expansive list of underlying foundational models. A Model represents a distinct neural network trained by various AI organizations (like OpenAI, Google, Anthropic, Meta, etc.).

Base Models

Base models are the standard foundational engines you chat with. They take a series of messages and output a sequence of text.

Naming Convention

Model names typically follow a prefix structure: <provider>/<model_name>. For example:

  • openai/gpt-4o-mini
  • anthropic/claude-3-haiku
  • meta-llama/llama-3.1-8b-instruct

Capabilities & Context Limits

Each model handles a different maximum “context limit” – the number of tokens (words/characters) it can process in a single request.

  • Fast, small models (e.g. llama-3) may have smaller limits but respond instantly.
  • Large models (e.g. gpt-4o) can handle massive documents and are heavily optimized for reasoning but take slightly longer to generate tokens.

Choosing the Right Model

When deciding what model to use for your application, consider these factors:

  • Cost: Do you need high intelligence, or just rapid categorization?
  • Speed (Latency): Lighter models offer much lower time-to-first-token.
  • Context Length: If you are passing an entire codebase or large PDF, ensure the model supports large contexts.

You can view the full dynamic list of supported models in our live Model Catalog.