ANTHROPIC MODEL FAMILY
Claude API Models: Sonnet, Opus, Fable and Haiku
Understand the Claude API model lineup, then choose Sonnet, Opus, Fable, or Haiku for coding, agents, long documents, latency, and cost.
Available models
Open a model page from this family
Read by decision
How to choose a Claude API model
Start with the smallest model likely to meet the workload. Compare it with one higher-capability candidate on the same tasks, then choose by accepted-result quality, latency, reliability, and total cost.
- Fable 5 targets the most demanding, long-running AI agent work.
- Opus 5 is the high-capability option for complex coding with AI agents and enterprise workflows.
- Sonnet 5 balances capability and speed for most production use; Haiku 4.5 is the faster, lower-cost option.
Claude API capabilities
- Fable 5, Opus 5 and Sonnet 5 publish a 1M context window and 128K maximum output.
- Haiku 4.5 publishes a 200K context window and 64K maximum output.
- The current top-tier models use adaptive thinking; Haiku exposes extended thinking.
- Claude is especially relevant for coding, document analysis and multi-step tool workflows.
Claude pricing and context questions
Do not compare only the headline input rate. Include output, cache reads and writes, long-context tiers, reasoning tokens, tool calls, retries, and the amount of history resent on every turn.
- Track model lifecycle as a deployment dependency; retired IDs can fail without a code change.
- For Claude Code, use the console-generated Anthropic-style environment configuration.
- Separate official Anthropic credentials from gateway credentials and verify the destination before long runs.
Which workloads fit Claude?
- Software development and repository maintenance with AI agents.
- Policy, contract, and research-document analysis.
- Long-running tool workflows with explicit checkpoints.
- Classification and support flows that need Haiku's lower latency.
Claude API limits and migration risks
- Lifecycle dates are operational deadlines, not optional release notes.
- OpenAI-compatible examples do not automatically apply to Claude Code or Anthropic-native clients.
- Validate tool schemas, streaming events and thinking controls on the exact LLMFly AI route.
Move from model research to a usable model ID
Open a model page above, note its provider model ID, and then use Model Plaza to confirm the ID available to your API key. Store that ID in configuration so it can be reviewed and changed without rewriting the application.
Frequently asked questions
Which Claude API model should I use?
Start with the smallest candidate whose published capabilities match the task, then compare it with one stronger model on a representative evaluation set.
How much does the Claude API cost?
Pricing belongs to a specific model. Open its page for provider pricing, then confirm the LLMFly AI price in Model Plaza.
What is the Claude context window?
Context limits vary by model. Use the specific model page and catalog instead of inferring a limit from the family name.
Can I use Claude through an OpenAI-compatible client?
Use a model marked compatible in Model Plaza and test the endpoint, streaming, tools, structured outputs, and error behavior your application needs.
Why keep the model ID in configuration?
It lets you test, roll back, and change models without scattering provider-specific IDs throughout the codebase.
Choose a Claude model
Open Model Plaza to confirm the model ID, API key access, availability, and current price.