# BestToken — AI Model Vendor Pricing Comparison

> Compare the same AI model across different API vendors to find the cheapest provider
> Data updated: 2026-05-15

## Pages

- [/models/text](/models/text) — LLM text model pricing (per 1M tokens)
- [/models/image](/models/image) — Image model pricing (per image)
- [/models/video](/models/video) — Video model pricing (per second)
- [/models/3d](/models/3d) — 3D model pricing (per generation)
- [/models/free](/models/free) — Free models & open weights
- [/services](/services) — AI service subscriptions

## Stats: 86 text models, 31 image models, 26 video models

## Vendors Compared

- Official API
- Azure OpenAI
- OpenRouter
- AWS Bedrock
- Google Vertex AI
- Together AI
- Fireworks AI
- SiliconFlow
- DeepInfra
- SambaNova
- 火山引擎
- 百度千帆
- Nebius AI
- Groq
- Cerebras
- Kimi
- 智谱 (GLM)

## FAQ

### Q: Where is the cheapest place to use GPT-4.1?
A: Official API: $2/M input, $8/M output | Azure OpenAI: $2/M input, $8/M output | OpenRouter: $2/M input, $8/M output

### Q: Where is the cheapest place to use DeepSeek V3?
A: Official API: $0.14/M input, $0.28/M output | Together AI: $0.14/M input, $0.28/M output | Fireworks AI: $0.14/M input, $0.28/M output | SiliconFlow: $0.14/M input, $0.28/M output | OpenRouter: $0.14/M input, $0.28/M output | DeepInfra: $0.14/M input, $0.28/M output | SambaNova: $0.14/M input, $0.28/M output | 火山引擎: $0.14/M input, $0.28/M output | 百度千帆: $0.14/M input, $0.28/M output

---
Website: https://www.uwa4d.cn