FeaturesPricingBlogFAQContact
Sign InGet Started
Price Comparison

LLM Price Per Token Comparison (2026)

Compare AI model pricing per token. 50+ models from OpenAI, Anthropic, Google, Meta, Mistral, and DeepSeek side by side. Click any column header to sort.

53 models
ModelProviderInput / 1M TokensOutput / 1M TokensContext WindowBest For
Mistral 7BMistral$0.03$0.0332KEdge deployment
Gemini 1.5 Flash-8BGoogle$0.04$0.151MUltra budget
Llama 3.1 8BMeta$0.05$0.05128KLightweight open
Llama 3 8BMeta$0.05$0.058KLegacy small
Gemma 2 9BGoogle$0.05$0.058KSmall open Google
Phi-4Meta$0.07$0.0716KSmall model tasks
Gemini 1.5 FlashGoogle$0.07$0.301MBudget long context
Gemini 2.0 FlashGoogle$0.10$0.401MBudget tasks
Mistral SmallMistral$0.10$0.30128KEfficient tasks
Cohere Embed v3Cohere$0.10$0.00512Embeddings
DeepSeek V2.5DeepSeek$0.14$0.28128KBudget general
DeepSeek Coder V2DeepSeek$0.14$0.28128KBudget code
Gemma 2 27BGoogle$0.14$0.148KOpen weight Google
GPT-4o-miniOpenAI$0.15$0.60128KHigh volume
Gemini 2.5 FlashGoogle$0.15$0.601MSpeed + value
Mistral NemoMistral$0.15$0.15128KOpen weight
Cohere Command RCohere$0.15$0.60128KEfficient RAG
Qwen 2.5 Coder 32BMeta$0.16$0.16128KCode open source
Llama 4 ScoutMeta$0.17$0.30512KLong context open
Llama 3.3 70BMeta$0.18$0.18128KOpen source
Llama 3.1 70BMeta$0.18$0.18128KBalanced open
Llama 4 MaverickMeta$0.19$0.49256KQuality open source
Qwen QwQ 32BMeta$0.20$0.20128KReasoning open
Llama 3 70BMeta$0.23$0.238KLegacy open
Mixtral 8x7BMistral$0.24$0.2432KMoE efficiency
Claude 3 HaikuAnthropic$0.25$1.25200KLegacy fast
DeepSeek V3DeepSeek$0.27$1.1064KCost-effective coding
Qwen 2.5 72BMeta$0.29$0.29128KMultilingual open
Mistral CodestralMistral$0.30$0.90256KCode generation
Cohere Command LightCohere$0.30$0.604KBudget Cohere
GPT-3.5 TurboOpenAI$0.50$1.5016KSimple tasks
DeepSeek R1DeepSeek$0.55$2.1964KReasoning
Mixtral 8x22BMistral$0.65$0.6564KLarge MoE
Claude Haiku 3.5Anthropic$0.80$4.00200KFast responses
Cohere CommandCohere$1.00$2.004KLegacy Cohere
o3-miniOpenAI$1.10$4.40200KReasoning at scale
o4-miniOpenAI$1.10$4.40200KEfficient reasoning
Gemini 2.5 ProGoogle$1.25$10.001MLong context
Gemini 1.5 ProGoogle$1.25$5.002MMax context
Mistral LargeMistral$2.00$6.00128KEnterprise
GPT-4oOpenAI$2.50$10.00128KGeneral purpose
Cohere Command R+Cohere$2.50$10.00128KRAG applications
Mistral MediumMistral$2.70$8.1032KLegacy enterprise
Claude Sonnet 4Anthropic$3.00$15.00200KBalanced performance
Claude 3 SonnetAnthropic$3.00$15.00200KLegacy balanced
Llama 3.1 405BMeta$3.00$3.00128KMax open source
GPT-5OpenAI$5.00$15.00200KComplex reasoning
o3OpenAI$10.00$40.00200KAdvanced reasoning
GPT-4 TurboOpenAI$10.00$30.00128KLegacy apps
Claude Opus 4Anthropic$15.00$75.00200KComplex analysis
Claude 3 OpusAnthropic$15.00$75.00200KLegacy complex
GPT-4OpenAI$30.00$60.008KLegacy compatibility
GPT-4 32KOpenAI$60.00$120.0032KLong legacy prompts

Track your actual spend across all these models

CostLayer monitors your real API usage across every provider and model, so you always know exactly what you are spending.

Get Started