Amux

Model comparison

Compare models side by side on price, context, performance, endpoints and features. Each column is one model on one provider — put the same model twice to compare providers.

Chat

Overview

DeveloperDeepSeek
DeveloperDeepSeek
Context1M token
Context1M token
Max output393K token
Max output393K token
ReasoningSupported
ReasoningSupported
Input
Input
Output
Output

Pricing

InputPeak$0.44/1M tokensOff-Peak$0.22/1M tokens
InputOff-Peak$0.22$0.187/1M tokensPeak$0.44$0.374/1M tokens
OutputPeak$1.32/1M tokensOff-Peak$0.66/1M tokens
OutputOff-Peak$0.66$0.561/1M tokensPeak$1.32$1.122/1M tokens
ReadPeak$0.014/1M tokensOff-Peak$0.007/1M tokens
ReadOff-Peak$0.022$0.0187/1M tokensPeak$0.044$0.0374/1M tokens

Performance

Avg. time to first token1.16 s
Avg. time to first token2.37 s
Avg. throughput124.7 tok/s
Avg. throughput88.6 tok/s
Availability100.0%
Availability100.0%

Averages over the recent window, weighted by request volume — not percentiles. Hours with no successful samples are left out rather than counted as zero.

Endpoints

OpenAI Chat Completions

/v1/chat/completions
/v1/chat/completions

OpenAI Responses

/v1/responses
/v1/responses

Anthropic Messages

/v1/messages
/v1/messages

Google Gemini

/v1beta/models/{model}:generateContent
/v1beta/models/{model}:generateContent

OpenAI Images

/v1/images/generations
/v1/images/generations
/v1/images/edits
/v1/images/edits

Alibaba Model Studio Video

/api/v1/services/aigc/video-generation/video-synthesis
/api/v1/services/aigc/video-generation/video-synthesis
/api/v1/tasks/{id}
/api/v1/tasks/{id}

MiniMax Video

/v2/video_generation
/v2/video_generation
/v2/query/video_generation/{id}
/v2/query/video_generation/{id}

Amux Tasks

/v1/tasks
/v1/tasks
/v1/tasks/{id}
/v1/tasks/{id}
/v1/tasks/{id}/cancel
/v1/tasks/{id}/cancel

Features

Streaming
Streaming
Reasoning
Reasoning
Function calling
Function calling
Structured output
Structured output
Prompt caching
Prompt caching
Vision
Vision
Image output
Image output
Video output
Video output