Cheapest AI models by price per million tokens

Ranked by combined input + output price per million tokens (excluding free-tier models). These are production-ready models that punch well above their price point — great defaults when cost matters and you can test model quality on your own workload.

1. 🥇
   
   meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo

   DeepInfra Inc.·$0.02 in / $0.05 out
   $0.03 avg
   $0.03 avg
2. 🥈
   
   meta-llama/llama-3-8b-instruct

   Novita AI·$0.04 in / $0.04 out
   $0.04 avg
   $0.04 avg
3. 🥉
   
   meta-llama/llama-3.2-3b-instruct

   Novita AI·$0.03 in / $0.05 out
   $0.04 avg
   $0.04 avg
4. 4
   
   sao10k/l3-8b-lunaris

   Novita AI·$0.05 in / $0.05 out
   $0.05 avg
   $0.05 avg
5. 5
   
   meta-llama/llama-3.1-8b-instruct

   Novita AI·$0.05 in / $0.05 out
   $0.05 avg
   $0.05 avg
6. 6
   
   Sao10K/L3-8B-Stheno-v3.2

   Novita AI·$0.05 in / $0.05 out
   $0.05 avg
   $0.05 avg
7. 7
   
   Qwen/Qwen3.5-2B

   DeepInfra Inc.·$0.02 in / $0.10 out
   $0.06 avg
   $0.06 avg
8. 8
   
   baichuan/baichuan-m2-32b

   Novita AI·$0.07 in / $0.07 out
   $0.07 avg
   $0.07 avg
9. 9
   
   Qwen/Qwen3-235B-A22B-Instruct-2507

   DeepInfra Inc.·$0.07 in / $0.10 out
   $0.09 avg
   $0.09 avg
10. 10
    
    gryphe/mythomax-l2-13b

    Novita AI·$0.09 in / $0.09 out
    $0.09 avg
    $0.09 avg
11. 11
    
    phi-4

    DeepInfra Inc.·$0.07 in / $0.14 out
    $0.10 avg
    $0.10 avg
12. 12
    
    gpt-5-nano:flex

    OpenAI Inc.·$0.02 in / $0.20 out
    $0.11 avg
    $0.11 avg
13. 13
    
    Qwen/Qwen2.5-Coder-32B-Instruct

    DeepInfra Inc.·$0.07 in / $0.16 out
    $0.12 avg
    $0.12 avg
14. 14
    
    nvidia/Nemotron-3-Nano-30B-A3B

    DeepInfra Inc.·$0.05 in / $0.20 out
    $0.13 avg
    $0.13 avg
15. 15
    
    qwen-turbo

    Alibaba Cloud·$0.05 in / $0.20 out
    $0.13 avg
    $0.13 avg
16. 16
    
    deepseek-ai/DeepSeek-V4-Flash

    DeepInfra Inc.·$0.10 in / $0.20 out
    $0.15 avg
    $0.15 avg
17. 17
    
    nvidia/nemotron-3-nano-omni

    Nebius AI·$0.06 in / $0.24 out
    $0.15 avg
    $0.15 avg
18. 18
    
    deepseek-r1-distill-qwen-14b

    Novita AI·$0.15 in / $0.15 out
    $0.15 avg
    $0.15 avg
19. 19
    
    mistralai/mistral-nemo

    Novita AI·$0.17 in / $0.17 out
    $0.17 avg
    $0.17 avg
20. 20
    gpt-oss-20b

    Fireworks AI·$0.07 in / $0.30 out
    $0.18 avg
    $0.18 avg
21. 21
    
    mistral-small-2503

    Mistral AI SAS·$0.10 in / $0.30 out
    $0.20 avg
    $0.20 avg
22. 22
    
    devstral-small-latest

    Mistral AI SAS·$0.10 in / $0.30 out
    $0.20 avg
    $0.20 avg
23. 23
    
    devstral-small-2507

    Mistral AI SAS·$0.10 in / $0.30 out
    $0.20 avg
    $0.20 avg
24. 24
    
    Qwen/Qwen3-32B

    DeepInfra Inc.·$0.10 in / $0.30 out
    $0.20 avg
    $0.20 avg
25. 25
    
    inclusionai/ling-2.6-flash

    Novita AI·$0.10 in / $0.30 out
    $0.20 avg
    $0.20 avg
26. 26
    
    xiaomimimo/mimo-v2-flash

    Novita AI·$0.10 in / $0.30 out
    $0.20 avg
    $0.20 avg
27. 27
    
    gemma-4-26B-A4B-it

    DeepInfra Inc.·$0.07 in / $0.34 out
    $0.20 avg
    $0.20 avg
28. 28
    deepseek-v4-flash

    Fireworks AI·$0.14 in / $0.28 out
    $0.21 avg
    $0.21 avg
29. 29
    
    meta-llama/Llama-3.3-70B-Instruct-Turbo

    DeepInfra Inc.·$0.12 in / $0.30 out
    $0.21 avg
    $0.21 avg
30. 30
    
    deepseek-chat

    DeepSeek·$0.14 in / $0.28 out
    $0.21 avg
    $0.21 avg