Models

TopRouter provides access to 200+ AI models. Use the API to list available models or browse the categories below.

List Models API

bash

curl https://toprouter.cc/models \
  -H "Authorization: Bearer sk-your-toprouter-key"

Model ID	Description
`anthropic/claude-4.8-opus`	Flagship Claude model for complex reasoning and agentic tasks
`anthropic/claude-4.6-sonnet`	Balanced performance and speed, primary daily driver
`anthropic/claude-4.5-haiku`	Fastest and most cost-effective Claude model

Model ID	Description
`openai/gpt-5.5`	Flagship multimodal GPT model, excellent at text and vision
`openai/gpt-5.5-instant`	High speed, cost-effective GPT-5.5 variant
`openai/gpt-5.5-thinking`	Advanced frontier reasoning model for complex logical tasks
`openai/gpt-5.4-thinking`	Standard reasoning model, balance of logic and response speed
`openai/o3`	Specialized reasoning model (deprecating August 2026)

Model ID	Description
`google/gemini-3.1-pro`	Most capable Gemini model for complex tasks and large contexts
`google/gemini-3.5-flash`	Fast, cost-effective Gemini model
`google/gemini-3.1-flash-lite`	Ultra-low latency, cost-efficient model

Model ID	Description
`deepseek/deepseek-v4-pro`	1.6T parameters flagship model, advanced coding and reasoning
`deepseek/deepseek-v4-flash`	High-speed, low-latency general purpose model

Model ID	Description
`openai/dall-e-3`	DALL-E 3 image generation
`openai/gpt-image-1`	GPT Image generation
`flux/flux-1-pro`	Flux professional image generation

Model ID	Description
`seedance/seedance-1.0`	AI video generation
`seedance/seedance-1-lit`	Lightweight AI video generation

INFO

Model availability and pricing may change. Use the /models endpoint for the most up-to-date list.

Cost-effective: Use google/gemini-3.5-flash for high-volume, general tasks
Best quality: Use anthropic/claude-4.8-opus for complex reasoning and analysis
Coding: Use anthropic/claude-4.6-sonnet or openai/gpt-5.5 for coding tasks
Fast responses: Use anthropic/claude-4.5-haiku or openai/gpt-5.5-instant
Image generation: Use openai/dall-e-3 or openai/gpt-image-1
Video generation: Use seedance/seedance-1.0