Skip to content

Cheapest models

Ranked by blended API price (75% input + 25% output, USD per 1M tokens). Free promotional listings excluded.

Ranked by blended API price (75% input + 25% output, USD per 1M tokens). Free promotional listings excluded. Source: OpenRouter. Variants (reasoning effort, batch) are folded into one row.

#ModelCapabilityReasoningCodingContext$/M
1Mistral NemoMistral AI118.629.9—131K$0.022
2Ling 3.0 Flash VLNEWinclusionAI (Ant Group)———262K$0.031
3Ling 3.0 FlashinclusionAI (Ant Group)———262K$0.032
4gpt-oss-20bOpenAI · +1 variant137.860.8—131K$0.036
5Granite 4.0 MicroIBM—28.3—131K$0.041
6Llama 3 8B LunarisSao10K———8K$0.043
7Nex-N2.5-MiniNEWNex AGI———262K$0.044
8gpt-oss-120bbatchOpenAI · +1 variant———131K$0.056
9Mistral Small 3Mistral AI127.147.3—33K$0.058
10Llama 3.1 8B InstructMeta———131K$0.058
11Schematron V2 TurboNEWInference.net———128K$0.060
12Gemma 3 4BGoogle116.023.2—131K$0.063
13Llama 3.2 1B InstructMeta———60K$0.070
14Gemma 3 12BGoogle123.539.5—131K$0.075
15Laguna XS 2.1Poolside———262K$0.075
16Ministral 3 8B 2512batchMistral AI · +1 variant———262K$0.075
17Hy-MT2-1.8BTencent———8K$0.077
18Qwen3 30B A3B Instruct 2507Alibaba (Qwen)———262K$0.084
19Nemotron 3.5 LightningNVIDIA———1M$0.085
20Phi 4Microsoft130.456.1—16K$0.087
21MythoMax 13Bgryphe———8K$0.087
22Nemotron 3 Nano 30B A3BNVIDIA———262K$0.087
23DeepSeek V4 Flash 0731DeepSeek154.591.0—1.31M$0.093
24GLM 5.3 FlashbatchZ.ai (Zhipu AI) · +1 variant———1.05M$0.095
25Schematron V2 SmallNEWInference.net———128K$0.095
26Ministral 3 3B 2512Mistral AI———131K$0.10
27Reka EdgeReka AI———16K$0.10
28Granite 4.2 8BNEWIBM———131K$0.11
29Qwen3.5-9BAlibaba (Qwen)139.479.0—262K$0.11
30Laguna S 2.1Poolside———1.05M$0.11
31Nex-N2.5-ProNEWNex AGI———262K$0.12
32Llama 3.2 3B InstructMeta———131K$0.12
33Qwen3 Coder 30B A3B InstructAlibaba (Qwen)———262K$0.12
34Qwen2.5 7B InstructAlibaba (Qwen)———33K$0.13
35Reka Flash 3Reka AI———66K$0.13
36UI-TARS 7BByteDance———128K$0.13
37Hy-MT2-30B-A3BTencent———8K$0.13
38Hy-MT2-7BTencent———8K$0.13
39Qwen3 32BAlibaba (Qwen)138.565.7—131K$0.13
40gpt-oss-safeguard-20bOpenAI———131K$0.13
41Mistral Small 4batchMistral AI · +1 variant———262K$0.13
42Mistral Small 3.2 24BMistral AI———256K$0.13
43Gemma 4 26B A4BGoogle141.873.2—262K$0.14
44GLM 4.7 FlashZ.ai (Zhipu AI)—60.5—200K$0.15
45Qwen3 14BAlibaba (Qwen)138.263.8—131K$0.15
46Llama 4 ScoutMeta129.651.8—1.31M$0.15
47Step 3.5 FlashStepFun———262K$0.15
48Voxtral Small 24B 2507Mistral AI———33K$0.15
49Gemma 4 31BGoogle142.775.8—262K$0.15
50Qwen3 235B A22B Instruct 2507Alibaba (Qwen)———262K$0.15
— means no published score from that source yet · click a model for every benchmark with its source

New - awaiting independent evaluation

All new models →

Not ranked until an independent evaluator (Epoch AI) publishes scores. We don't use vendor-reported benchmark claims for rankings.

Best AI for coding →
Models ranked on DeepSWE
Best AI for reasoning →
Ranked on GPQA Diamond
Best AI for math →
Ranked on competition math
Best open-weight models →
Weights you can run yourself
Cheapest capable models →
Lowest blended API price
Best AI coding tools →
Apps & agents by popularity
Best AI image generators →
Apps & models
Best AI for research →
Answer engines & deep research