LLM API Pricing Comparison 2026

Compare input and output token pricing across 92 large language models from OpenAI, Anthropic, Google, Meta, and more. Sort by any column, filter by provider or capability, and click any model to see full benchmarks and details.

Data verified Apr 20, 2026
Capabilities:
Last verified: Apr 20, 2026
Showing 92 of 92 models
ModelProviderInput $/MOutput $/MArena ELO
Claude Opus 4Anthropic$5.00$25.001503
Gemini 2.5 ProGoogle$1.25$10.001430
o3OpenAI$2.00$8.001340
DeepSeek R1DeepSeek$0.500$2.151310
o1OpenAI$15.00$60.001310
Qwen 3 235B MoEAlibaba$0.455$1.821310
Gemini Experimental 1206Google$0.00$0.001300
GPT-4.5OpenAI$75.00$150.001290
DeepSeek R1 (Groq)Groq$0.750$0.9901290
Llama 4 MaverickMeta$0.150$0.6001290
DeepSeek R1 (Together)Together AI$3.00$7.001290
Grok 3xAI$3.00$15.001285
Claude Sonnet 4Anthropic$3.00$15.001280
DeepSeek V3DeepSeek$0.259$0.4201280
Gemini 2.0 Flash ThinkingGoogle$0.00$0.001280
o3-miniOpenAI$1.10$4.401280
Claude 3.5 SonnetAnthropic$3.00$15.001270
Gemini 2.5 FlashGoogle$0.300$2.501270
o1-miniOpenAI$1.10$4.401270
ChatGPT-4o LatestOpenAI$5.00$15.001265
Gemini 2.0 FlashGoogle$0.100$0.4001260
GPT-4oOpenAI$2.50$10.001260
o4-miniOpenAI$1.10$4.401260
Qwen 2.5 MaxAlibaba$0.160$0.6401260
QwQ 32BAlibaba$0.150$0.5801260
GPT-4o (Aug 2024)OpenAI$2.50$10.001255
DeepSeek R1 Distill Llama 70BDeepSeek$0.700$0.8001250
Llama 4 ScoutMeta$0.080$0.3001250
Mistral LargeMistral$0.500$1.501245
Command ACohere$2.50$10.001240
DeepSeek R1 Distill Qwen 32BDeepSeek$0.290$0.2901240
Llama 3.1 405B (Fireworks)Fireworks AI$3.00$3.001240
GPT-4 TurboOpenAI$10.00$30.001240
Grok 2xAI$2.00$10.001240
Llama 3.1 405BMeta$3.00$3.001240
Sonar ReasoningPerplexity$2.00$8.001240
Llama 3.1 405B (Together)Together AI$3.50$3.501240
Gemini 1.5 ProGoogle$1.25$5.001230
Grok 2 VisionxAI$2.00$10.001230
Pixtral LargeMistral AI$2.00$6.001230
Qwen 2.5 72BAlibaba$0.120$0.3901230
Qwen 2.5 72B (Together)Together AI$1.20$1.201230
Amazon Nova ProAmazon$0.800$3.201220
Claude 3.5 HaikuAnthropic$0.800$4.001220
Claude Haiku 4Anthropic$1.00$5.001220
Llama 3.3 70B (Fireworks)Fireworks AI$0.900$0.9001220
GPT-4o MiniOpenAI$0.150$0.6001220
Llama 3.3 70B (Groq)Groq$0.590$0.7901220
Llama 3.3 70BMeta$0.120$0.3801220
Mistral Medium 3Mistral AI$0.400$2.001220
Llama 3.3 70B (Together)Together AI$0.880$0.8801220
Llama 3.2 90B VisionMeta$0.900$0.9001210
Command R+Cohere$2.50$10.001200
DeepSeek V2.5DeepSeek$0.140$0.2801200
Mixtral 8x22B (Fireworks)Fireworks AI$0.900$0.9001200
Gemini 2.0 Flash LiteGoogle$0.075$0.3001200
GPT-4 1OpenAI$2.00$8.001200
Sonar ProPerplexity$3.00$15.001200
WizardLM-2 8x22BMicrosoft$0.620$0.6201200
Llama 3.1 70BMeta$0.400$0.4001195
Phi-3.5 MoEMicrosoft$0.170$0.6801195
Gemini 1.5 FlashGoogle$0.075$0.3001190
Gemma 2 27BGoogle$0.650$0.6501190
Mistral SmallMistral$0.150$0.6001185
Yi-Large01.AI$3.00$3.001185
GPT-4 1.5-miniOpenAI$0.400$1.601180
Grok 3-minixAI$0.300$0.5001175
Amazon Nova LiteAmazon$0.060$0.2401170
Gemma 2 9B (Groq)Groq$0.200$0.2001170
Phi-3 MediumMicrosoft$0.170$0.1701170
Yi-Lightning01.AI$0.140$0.1401165
Gemma 2 9BGoogle$0.030$0.0901160
Mixtral 8x7B (Groq)Groq$0.240$0.2401160
Llama 3.2 11B VisionMeta$0.245$0.2451160
Phi-3.5 MiniMicrosoft$0.130$0.5201160
Qwen 2.5 7BAlibaba$0.040$0.1001160
SonarPerplexity$1.00$1.001160
InternLM 2.5 20BShanghai AI Lab$0.180$0.1801155
Gemini 1.5 Flash 8BGoogle$0.037$0.1501150
GPT-4 1.5-nanoOpenAI$0.100$0.4001150
Phi-4Microsoft$0.065$0.1401150
Command RCohere$0.150$0.6001140
Mistral Nemo 12BMistral AI$0.020$0.0401140
Amazon Nova MicroAmazon$0.035$0.1401130
Command R7BCohere$0.038$0.1501120
GPT-3.5 TurboOpenAI$0.500$1.501120
Llama 3.1 8B (Groq)Groq$0.050$0.0801120
Llama 3.1 8BMeta$0.020$0.0501120
Mistral 7BMistral AI$0.110$0.1901100
Mistral 7B (Together)Together AI$0.200$0.2001100
Codestral 22BMistral AI$0.300$0.900--
Qwen 2.5 Coder 32BAlibaba$0.660$1.00--

Frequently Asked Questions

Which LLM API is the cheapest in 2026?
As of April 2026, GPT-4.1 Nano and Gemini 2.0 Flash Lite offer the lowest per-token pricing for production workloads. Prices vary by input vs. output tokens, so the cheapest option depends on your specific usage pattern.
How often are LLM API prices updated?
We verify pricing directly from provider documentation every week. Each model listing shows a 'Last verified' date so you can confirm the data is current.
What is the difference between input and output token pricing?
Input tokens are the tokens you send to the API (your prompt), while output tokens are the tokens the model generates in its response. Most providers charge different rates for each, with output tokens typically costing 2-5x more than input tokens.
Do any LLM APIs offer free tiers?
Several providers offer limited free tiers or trial credits. Google's Gemini API has a generous free tier for lower rate limits. OpenAI and Anthropic offer sign-up credits for new accounts. Check each provider's pricing page for current free tier details.