Cheapest LLM APIs
The lowest-cost large language model APIs we track, ranked by combined input and output price per million tokens. Most of the cheapest options are open-weight models (DeepSeek, Qwen, GLM, MiniMax) served by competing providers — which is exactly why prices keep falling.
Ranking 50 options · updated November 22, 2026
| # | Model | Provider | Input /1M | Output /1M |
|---|---|---|---|---|
| 1 | Llama 3.2 1B Instruct open | DeepInfra | $0.0050 | $0.010 |
| 2 | Llama 3.2 1B Instruct open | Novita AI | $0.020 | $0.020 |
| 3 | Llama 3.2 3B Instruct open | DeepInfra | $0.020 | $0.020 |
| 4 | DeepSeek V4 Flash open | OpenRouter | $0.015 | $0.029 |
| 5 | Llama 3.1 8B Instruct open | OpenRouter | $0.020 | $0.040 |
| 6 | Llama 3.1 8B Instruct open | Novita AI | $0.020 | $0.050 |
| 7 | Qwen3.7 Flash | Infron | $0.017 | $0.066 |
| 8 | Gemma 3 4B open | OpenRouter | $0.040 | $0.050 |
| 9 | gpt-oss-20b open | OpenRouter | $0.017 | $0.085 |
| 10 | gpt-oss-20b open | Synap by Linkrra | $0.017 | $0.085 |
| 11 | gpt-oss-20b open | AkashML | $0.018 | $0.090 |
| 12 | Nemotron 3.5 Lightning open | OpenRouter | $0.020 | $0.096 |
| 13 | GLM 5.3 Flash open | OpenRouter | $0.040 | $0.078 |
| 14 | Llama 3.1 8B Instruct open | WaveSpeed AI | $0.050 | $0.080 |
| 15 | Solar Pro 4 | Infron | $0.030 | $0.120 |
| 16 | Gemma 3 4B open | DeepInfra | $0.050 | $0.100 |
| 17 | Gemma 3 12B open | Novita AI | $0.050 | $0.100 |
| 18 | Gemma 3 4B open | Infron | $0.050 | $0.100 |
| 19 | Qwen3.7 Flash | OpenRouter | $0.030 | $0.130 |
| 20 | Qwen3.7 Flash | Alibaba Cloud | $0.030 | $0.130 |
| 21 | Qwen3.7 Flash | WaveSpeed AI | $0.030 | $0.130 |
| 22 | Qwen3.7 Flash | Token Harbor | $0.030 | $0.130 |
| 23 | Gemma 3 4B open | Runcrate | $0.055 | $0.110 |
| 24 | gpt-oss-20b open | DeepInfra | $0.030 | $0.140 |
| 25 | gpt-oss-20b open | Infron | $0.030 | $0.140 |
| 26 | Hy4 Flash open | OpenRouter | $0.035 | $0.141 |
| 27 | Hy4 Flash open | Novita AI | $0.035 | $0.141 |
| 28 | DeepSeek V4 Flash open | OpenRouter | $0.044 | $0.132 |
| 29 | gpt-oss-20b open | Parasail | $0.030 | $0.150 |
| 30 | GPT-6 Luna | KIE AI | $0.030 | $0.150 |
| 31 | GPT OSS 120B open | OpenRouter | $0.030 | $0.150 |
| 32 | Hy4 Flash open | Infron | $0.040 | $0.140 |
| 33 | gpt-oss-20b open | Runcrate | $0.033 | $0.154 |
| 34 | gpt-oss-20b open | Novita AI | $0.040 | $0.150 |
| 35 | Qwen 3.5 9B open | Infron | $0.040 | $0.150 |
| 36 | Gemma 3 12B open | OpenRouter | $0.050 | $0.150 |
| 37 | Gemma 3 12B open | DeepInfra | $0.050 | $0.150 |
| 38 | Gemma 3 12B open | Infron | $0.050 | $0.150 |
| 39 | GPT OSS 120B open | DeepInfra | $0.037 | $0.170 |
| 40 | GPT OSS 120B open | Infron | $0.037 | $0.170 |
| 41 | Granite 4.2 8B open | OpenRouter | $0.060 | $0.150 |
| 42 | Qwen 3.5 9B open | OpenRouter | $0.080 | $0.130 |
| 43 | Nemotron 3.5 Lightning open | DeepInfra | $0.060 | $0.160 |
| 44 | Gemma 3 12B open | Runcrate | $0.055 | $0.165 |
| 45 | GPT OSS 120B open | AkashML | $0.037 | $0.187 |
| 46 | Llama 3.2 1B Instruct open | Cloudflare Workers AI | $0.027 | $0.201 |
| 47 | Llama 3.2 1B Instruct open | OpenRouter | $0.027 | $0.201 |
| 48 | GPT OSS 120B open | Runcrate | $0.041 | $0.187 |
| 49 | Gemma 3 27B open | OpenRouter | $0.080 | $0.160 |
| 50 | Gemma 4 26B A4B open | OpenRouter | $0.050 | $0.190 |