Best cheap AI models
Capable models only (overall index of 140 or more in independent tests), ranked from cheapest to most expensive using a typical mix of 3 input tokens for 1 output token.
Our picks
The ranking
Compare the top two| # | Model | Blended price | Overall indexi | Inputi | Outputi | Contexti |
|---|---|---|---|---|---|---|
| 1 | Qwen3.7 FlashAlibaba Qwen | $0.055 | 144.6 | $0.03 | $0.13 | 1M |
| 2 | DeepSeek V4 Flash 0731DeepSeek | $0.094 | 154.5 | $0.018 | $0.32 | 1M |
| 3 | Qwen3.5-FlashAlibaba Qwen | $0.114 | 144.0 | $0.065 | $0.26 | 1M |
| 4 | Gemma 4 26B A4B Google DeepMind | $0.121 | 141.9 | $0.077 | $0.26 | 256K |
| 5 | Gemma 4 31BGoogle DeepMind | $0.153 | 142.7 | $0.09 | $0.34 | 256K |
| 6 | GLM 5.3 FlashZ.ai (Zhipu) | $0.238 | 151.9 | $0.15 | $0.50 | 1M |
| 7 | DeepSeek V3.2 ExpDeepSeek | $0.305 | 145.0 | $0.27 | $0.41 | 160K |
| 8 | DeepSeek V3.2DeepSeek | $0.315 | 146.3 | $0.28 | $0.42 | 160K |
| 9 | Qwen3.6 35B A3BAlibaba Qwen | $0.363 | 143.9 | $0.15 | $1 | 256K |
| 10 | MiniMax M2.7MiniMax | $0.368 | 145.8 | $0.21 | $0.84 | 200K |
per 1M tokens · USD · Blended price per million tokens, for a typical mix of 3 input tokens to 1 output token.
Quick answers
- What is the best cheap AI model?
- Qwen3.7 Flash from Alibaba Qwen leads this ranking (Blended price: $0.055). It costs $0.03 per million input tokens and $0.13 per million output tokens.
How we rank
Scores come from tests run independently by Epoch AI (CC BY 4.0), not from the companies’ own announcements. Prices come from OpenRouter’s public API. Both are refreshed automatically several times a day, so this ranking follows new releases and price changes.
Other rankings
CodingMathScienceFreeOpen sourceLong contextImages & PDFsRankings →