The cheapest LLM APIs

The cheapest LLM API right now is ministral-3-3b-2512 from Mistral: $0.1 per million input tokens and $0.1 per million output tokens, so a typical request of 1,000 input and 500 output tokens costs $0.00015. Below are the 20 cheapest of 86 models, and the cheapest from each provider.

Prices updated November 14, 2026

Cheapest models overall

ModelProviderInput / 1MCached input / 1MOutput / 1MTypical request
Ministral 3 3B 2512Mistral$0.1$0.01$0.1$0.00015
Ministral 3B 2512Mistral$0.1$0.01$0.1$0.00015
Ministral 3 8B 2512Mistral$0.15$0.015$0.15$0.000225
Ministral 8B 2512Mistral$0.15$0.015$0.15$0.000225
GPT-5 nanoOpenAI$0.05$0.005$0.4$0.00025
Mistral SmallMistral$0.1$0.01$0.3$0.00025
DeepSeek CoderDeepSeek$0.14$0.014$0.28$0.00028
Gemini 2.5 Flash LiteGoogle Gemini$0.1$0.01$0.4$0.0003
Ministral 3 14B 2512Mistral$0.2$0.02$0.2$0.0003
Ministral 14B 2512Mistral$0.2$0.02$0.2$0.0003
GPT-6 LunaOpenAI$0.1$0.01$0.5$0.00035
Claude Haiku 5.5Anthropic$0.1$0.01$0.5$0.00035
Mistral TinyMistral$0.25$0.025$0.25$0.000375
GPT-4o miniOpenAI$0.15$0.075$0.6$0.00045
Open Mistral NemoMistral$0.3$0.03$0.3$0.00045
DeepSeek V3.2DeepSeek$0.28$0.028$0.4$0.00048
Codestral 2508Mistral$0.3$0.03$0.9$0.00075
GPT-5.6 LunaOpenAI$0.2$0.02$1.2$0.0008
DeepSeek V3DeepSeek$0.27$0.07$1.1$0.00082
GPT-5.4 nanoOpenAI$0.2$0.02$1.25$0.000825

Cheapest model from each provider

ModelProviderInput / 1MCached input / 1MOutput / 1MTypical request
GPT-5 nanoOpenAI$0.05$0.005$0.4$0.00025
Claude Haiku 5.5Anthropic$0.1$0.01$0.5$0.00035
Gemini 2.5 Flash LiteGoogle Gemini$0.1$0.01$0.4$0.0003
Ministral 3 3B 2512Mistral$0.1$0.01$0.1$0.00015
DeepSeek CoderDeepSeek$0.14$0.014$0.28$0.00028
Grok Build 0.1xAI$1$0.2$2$0.002

Cheapest model with a given feature

FeatureModelProviderInput / 1MOutput / 1MTypical request
Reasoninggpt-5-nanoOpenAI$0.05$0.4$0.00025
Image inputministral-3-3b-2512Mistral$0.1$0.1$0.00015
Tool callingministral-3-3b-2512Mistral$0.1$0.1$0.00015
1M+ token contextgemini-2.5-flash-liteGoogle Gemini$0.1$0.4$0.0003

Ranked by the cost of a typical request of 1,000 input and 500 output tokens at standard rates. Models listed at $0 are left out. The batch API and prompt caching make most models cheaper still, see batch API pricing and prompt caching pricing.

Calculate the cost of your own prompt →

Question and answers

What is the cheapest LLM API?

The cheapest LLM API right now is ministral-3-3b-2512 from Mistral: $0.1 per million input tokens and $0.1 per million output tokens, so a typical request of 1,000 input and 500 output tokens costs $0.00015.