Home › Cheapest LLM APIs

Cheapest LLM APIs

The cheapest paid LLM API is IBM: Granite 4.0 Micro at $0.017 per 1M input tokens and $0.112 output. 20 models are completely free at list price. Rankings below cover the 25 cheapest models by input price, output price, and blended (80% input / 20% output) price, from the October 21, 2026 snapshot.

Cheapest by input price

#ModelProviderInput /1MOutput /1MContext
1MiniMax: MiniMax M3.1 Flash (free)minimaxFreeFree1,000,000
2XingChen: Xing4.0 29B A4B (free)xingchen-agiFreeFree262,144
3NVIDIA: Nemotron 3.5 Super VL (free)nvidiaFreeFree1,000,000
4Dots Studio: Dots3-Note Preview (free)dots-studioFreeFree512,000
5LiquidAI: LFM2.5-2.6B (free)liquidFreeFree65,536
6NVIDIA: Nemotron 3.5 Lightning (free)nvidiaFreeFree1,000,000
7Thinking Machines: Inkling Small (free)thinkingmachinesFreeFree1,048,576
8Poolside: Laguna S 2.1 (free)poolsideFreeFree262,144
9Thinking Machines: Inkling (free)thinkingmachinesFreeFree1,048,576
10Poolside: Laguna XS 2.1 (free)poolsideFreeFree262,144
11Cohere: North Mini Code (free)cohereFreeFree256,000
12NVIDIA: Nemotron 3.5 Content Safety (free)nvidiaFreeFree128,000
13NVIDIA: Nemotron 3 Ultra (free)nvidiaFreeFree1,000,000
14NVIDIA: Nemotron 3 Nano Omni (free)nvidiaFreeFree256,000
15Google: Gemma 4 26B A4B (free)googleFreeFree262,144
16Google: Gemma 4 31B (free)googleFreeFree262,144
17Google: Lyria 3 Pro PreviewgoogleFreeFree1,048,576
18Google: Lyria 3 Clip PreviewgoogleFreeFree1,048,576
19NVIDIA: Nemotron 3 Super (free)nvidiaFreeFree262,144
20Free Models RouteropenrouterFreeFree200,000
21IBM: Granite 4.0 Microibm-granite$0.017$0.112131,000
22OpenAI: gpt-oss-20bopenai$0.018$0.09131,072
23inclusionAI: Ling 3.0 Flashinclusionai$0.021$0.063262,144
24DeepSeek: DeepSeek V4.1 Flashdeepseek$0.02394$0.3751,048,576
25OpenAI: gpt-oss-20b (batch)openai$0.024$0.112131,072

Cheapest by output price

#ModelProviderInput /1MOutput /1MContext
1MiniMax: MiniMax M3.1 Flash (free)minimaxFreeFree1,000,000
2XingChen: Xing4.0 29B A4B (free)xingchen-agiFreeFree262,144
3NVIDIA: Nemotron 3.5 Super VL (free)nvidiaFreeFree1,000,000
4Dots Studio: Dots3-Note Preview (free)dots-studioFreeFree512,000
5LiquidAI: LFM2.5-2.6B (free)liquidFreeFree65,536
6NVIDIA: Nemotron 3.5 Lightning (free)nvidiaFreeFree1,000,000
7Thinking Machines: Inkling Small (free)thinkingmachinesFreeFree1,048,576
8Poolside: Laguna S 2.1 (free)poolsideFreeFree262,144
9Thinking Machines: Inkling (free)thinkingmachinesFreeFree1,048,576
10Poolside: Laguna XS 2.1 (free)poolsideFreeFree262,144
11Cohere: North Mini Code (free)cohereFreeFree256,000
12NVIDIA: Nemotron 3.5 Content Safety (free)nvidiaFreeFree128,000
13NVIDIA: Nemotron 3 Ultra (free)nvidiaFreeFree1,000,000
14NVIDIA: Nemotron 3 Nano Omni (free)nvidiaFreeFree256,000
15Google: Gemma 4 26B A4B (free)googleFreeFree262,144
16Google: Gemma 4 31B (free)googleFreeFree262,144
17Google: Lyria 3 Pro PreviewgoogleFreeFree1,048,576
18Google: Lyria 3 Clip PreviewgoogleFreeFree1,048,576
19NVIDIA: Nemotron 3 Super (free)nvidiaFreeFree262,144
20Free Models RouteropenrouterFreeFree200,000
21ScaleDown: Summarizescaledown$0.04Free1,000,000
22ScaleDown: Extractscaledown$0.04Free1,000,000
23Mistral: Mistral Nemomistralai$0.029$0.03131,072
24Google: Gemma 3 4Bgoogle$0.04$0.05131,072
25Sao10K: Llama 3 8B Lunarissao10k$0.04$0.058,192

Cheapest by blended price (80/20)

#ModelProviderInput /1MOutput /1MContext
1MiniMax: MiniMax M3.1 Flash (free)minimaxFreeFree1,000,000
2XingChen: Xing4.0 29B A4B (free)xingchen-agiFreeFree262,144
3NVIDIA: Nemotron 3.5 Super VL (free)nvidiaFreeFree1,000,000
4Dots Studio: Dots3-Note Preview (free)dots-studioFreeFree512,000
5LiquidAI: LFM2.5-2.6B (free)liquidFreeFree65,536
6NVIDIA: Nemotron 3.5 Lightning (free)nvidiaFreeFree1,000,000
7Thinking Machines: Inkling Small (free)thinkingmachinesFreeFree1,048,576
8Poolside: Laguna S 2.1 (free)poolsideFreeFree262,144
9Thinking Machines: Inkling (free)thinkingmachinesFreeFree1,048,576
10Poolside: Laguna XS 2.1 (free)poolsideFreeFree262,144
11Cohere: North Mini Code (free)cohereFreeFree256,000
12NVIDIA: Nemotron 3.5 Content Safety (free)nvidiaFreeFree128,000
13NVIDIA: Nemotron 3 Ultra (free)nvidiaFreeFree1,000,000
14NVIDIA: Nemotron 3 Nano Omni (free)nvidiaFreeFree256,000
15Google: Gemma 4 26B A4B (free)googleFreeFree262,144
16Google: Gemma 4 31B (free)googleFreeFree262,144
17Google: Lyria 3 Pro PreviewgoogleFreeFree1,048,576
18Google: Lyria 3 Clip PreviewgoogleFreeFree1,048,576
19NVIDIA: Nemotron 3 Super (free)nvidiaFreeFree262,144
20Free Models RouteropenrouterFreeFree200,000
21Mistral: Mistral Nemomistralai$0.029$0.03131,072
22inclusionAI: Ling 3.0 Flashinclusionai$0.021$0.063262,144
23ScaleDown: Summarizescaledown$0.04Free1,000,000
24ScaleDown: Extractscaledown$0.04Free1,000,000
25OpenAI: gpt-oss-20bopenai$0.018$0.09131,072

Ranking from the October 21, 2026 snapshot, refreshed daily. How we collect prices