Skip to main content
Lightweight gives you access to 25+ models from four providers through a single API key. Every model works with the same POST /v1/chat/completions endpoint.
For the live, always-up-to-date catalog, call GET /v1/models. See the Models API reference for full details.

​
OpenAI

ModelContext WindowStatus
gpt-5.4128KAvailable
gpt-5.4-mini128KAvailable
gpt-5.2128KAvailable
gpt-5.11MAvailable
gpt-5.1-codex1MAvailable
gpt-5-mini128KAvailable
gpt-4.11MAvailable
gpt-4.1-mini1MAvailable
gpt-4.1-nano1MAvailable
gpt-4o128KAvailable
gpt-4o-mini128KAvailable
o3200KAvailable
o3-mini200KAvailable
o4-mini200KAvailable

​
Anthropic

ModelContext WindowStatus
claude-opus-4.6200KAvailable
claude-opus-4.6-fast200KAvailable
claude-sonnet-4.6200KAvailable
claude-sonnet-4.5200KAvailable
claude-opus-4.5200KAvailable
claude-haiku-4.5200KAvailable

​
Google

ModelContext WindowStatus
gemini-3.1-pro1MAvailable
gemini-2.5-pro1MAvailable
gemini-3-flash1MAvailable

​
xAI

ModelContext WindowStatus
grok-code-fast-1128KAvailable

​
Model Aliases

For convenience, Lightweight supports short alias names that resolve to the latest version:
AliasResolves To
claude-opusclaude-opus-4.6
claude-sonnetclaude-sonnet-4.6
claude-haikuclaude-haiku-4.5
gptgpt-5.4
gemini-progemini-3.1-pro
gemini-flashgemini-3-flash
Aliases always point to the latest stable version. If you need deterministic behavior, use the full model name (e.g. claude-opus-4.6 instead of claude-opus).

​
Extended Thinking

Anthropic Claude models support extended thinking (chain-of-thought) when you pass the appropriate parameter. This allows the model to reason through complex problems step by step before responding. Extended thinking is supported on:
  • claude-opus-4.6
  • claude-opus-4.6-fast
  • claude-sonnet-4.6
  • claude-sonnet-4.5
See the Anthropic SDK integration guide for usage examples.

​
Live Catalog

The model list above is a point-in-time snapshot. For the always-current catalog, use the API: 
curl https://api.lightweight.one/v1/models \
  -H "Authorization: Bearer lw_sk_your-key-here"
See Models API for full response schema and filtering options.