LLM API pricing
LLM API pricing, every vendor side by side
147 model price rows from 21 vendors, each read off the vendor's own pricing page with a link on every row. Sort the table, then estimate a monthly bill across all of them.
21
Vendors
142
Models priced
$0.03
Lowest input / 1M
2026-10-25
Prices checked
Prices
Every model, sorted by price
Standard prices in USD per 1M tokens. Click a column to sort. DeepSeek rows are peak prices; off-peak is half.
| Source | ||||||
|---|---|---|---|---|---|---|
| qwen3.7-flashprompts up to 32K | Alibaba Cloud | $0.03 | none listed | $0.13 | $0.065 | Sourcechecked 2026-10-25 |
| Amazon Nova Microus-east-1, in-region | Amazon Web Services | $0.035 | $0.00875 | $0.14 | $0.07 | Sourcechecked 2026-10-25 |
| Command R7B | Cohere | $0.0375 | none listed | $0.15 | none listed | Sourcechecked 2026-10-25 |
| NVIDIA Nemotron 3.5 Lightning 30B A3B | Fireworks AI | $0.05 | $0.01 | $0.20 | $0.10 | Sourcechecked 2026-10-25 |
| qwen-flashprompts up to 256K | Alibaba Cloud | $0.05 | none listed | $0.40 | $0.20 | Sourcechecked 2026-10-25 |
| GPT OSS 20B | Groq | $0.075 | $0.037 | $0.30 | $0.15 | Sourcechecked 2026-10-25 |
| Safety GPT OSS 20BPreview | Groq | $0.075 | none listed | $0.30 | $0.15 | Sourcechecked 2026-10-25 |
| Anthropic: Claude Haiku 5.5 | OpenRouter | $0.10 | $0.01 | $0.50 | none listed | Sourcechecked 2026-10-25 |
| Claude Haiku 5.5prompts up to 100,000 tokens | Anthropic | $0.10 | $0.01 | $0.50 | $0.25 | Sourcechecked 2026-10-25 |
| Claude Haiku 5.5prompts up to 200K | Google Cloud | $0.10 | $0.01 | $0.50 | $0.25 | Sourcechecked 2026-10-25 |
| Gemini 2.5 Flash-Litetext, image and video input | $0.10 | $0.01 | $0.40 | $0.20 | Sourcechecked 2026-10-25 | |
| gpt-6-luna | OpenAI | $0.10 | $0.01 | $0.50 | $0.25 | Sourcechecked 2026-10-25 |
| Ministral 3 3B | Mistral AI | $0.10 | none listed | $0.10 | $0.05 | Sourcechecked 2026-10-25 |
| OpenAI: GPT-6 Luna | OpenRouter | $0.10 | $0.01 | $0.50 | none listed | Sourcechecked 2026-10-25 |
| Claude Haiku 5.5us-east-1, in-region | Amazon Web Services | $0.11 | $0.011 | $0.55 | none listed | Sourcechecked 2026-10-25 |
| Command R | Cohere | $0.15 | none listed | $0.60 | none listed | Sourcechecked 2026-10-25 |
| DeepSeek: DeepSeek V4.1 Flash 1017 | OpenRouter | $0.15 | $0.003 | $0.60 | none listed | Sourcechecked 2026-10-25 |
| GLM 5.3 Flash | Fireworks AI | $0.15 | $0.03 | $0.50 | $0.25 | Sourcechecked 2026-10-25 |
| GLM-5.3-Flash | Together AI | $0.15 | $0.03 | $0.50 | none listed | Sourcechecked 2026-10-25 |
| GPT OSS 120B | Groq | $0.15 | $0.075 | $0.60 | $0.30 | Sourcechecked 2026-10-25 |
| gpt-oss-120bus-east-1, in-region | Amazon Web Services | $0.15 | none listed | $0.60 | $0.30 | Sourcechecked 2026-10-25 |
| gpt-oss-120B | Together AI | $0.15 | none listed | $0.60 | none listed | Sourcechecked 2026-10-25 |
| MiniMax-M3.1-Flash | MiniMax | $0.15 | $0.015 | $0.60 | none listed | Sourcechecked 2026-10-25 |
| Ministral 3 8B | Mistral AI | $0.15 | none listed | $0.15 | $0.075 | Sourcechecked 2026-10-25 |
| Mistral Small 4 | Mistral AI | $0.15 | none listed | $0.60 | $0.30 | Sourcechecked 2026-10-25 |
| OpenAI GPT OSS 120B | Fireworks AI | $0.15 | $0.015 | $0.60 | $0.30 | Sourcechecked 2026-10-25 |
| OpenAI: gpt-oss-120b | OpenRouter | $0.15 | $0.075 | $0.60 | none listed | Sourcechecked 2026-10-25 |
| perplexity/glm-5.3-flash | Perplexity | $0.15 | $0.03 | $0.50 | none listed | Sourcechecked 2026-10-25 |
| qwen3.8-flash | Alibaba Cloud | $0.15 | none listed | $0.47 | none listed | Sourcechecked 2026-10-25 |
| Qwen3.5 9B | Together AI | $0.17 | none listed | $0.25 | none listed | Sourcechecked 2026-10-25 |
| openai/gpt-oss-120b | Replicate | $0.18 | none listed | $0.72 | none listed | Sourcechecked 2026-10-25 |
| GPT-5.4 nanoGlobal | Microsoft Azure | $0.20 | $0.02 | $1.25 | none listed | Sourcechecked 2026-10-25 |
| gpt-5.4-nano | OpenAI | $0.20 | $0.02 | $1.25 | $0.625 | Sourcechecked 2026-10-25 |
| gpt-5.6-luna | OpenAI | $0.20 | $0.02 | $1.20 | $0.60 | Sourcechecked 2026-10-25 |
| GPT-5.6-lunaGlobal, short context | Microsoft Azure | $0.20 | $0.02 | $1.20 | none listed | Sourcechecked 2026-10-25 |
| Ministral 3 14B | Mistral AI | $0.20 | none listed | $0.20 | $0.10 | Sourcechecked 2026-10-25 |
| perplexity/nemotron-3.5-super-120b-a12b | Perplexity | $0.20 | $0.04 | $0.50 | none listed | Sourcechecked 2026-10-25 |
| Llama 4 Maverick 17Bus-east-1, in-region | Amazon Web Services | $0.24 | none listed | $0.97 | $0.485 | Sourcechecked 2026-10-25 |
| Gemini 3.1 Flash-Litetext, image and video input | $0.25 | $0.025 | $1.50 | $0.75 | Sourcechecked 2026-10-25 | |
| perplexity/sonar | Perplexity | $0.25 | $0.0625 | $2.50 | none listed | Sourcechecked 2026-10-25 |
| Codestral | Mistral AI | $0.30 | none listed | $0.90 | $0.45 | Sourcechecked 2026-10-25 |
| DeepSeek V4.1 Flash | Fireworks AI | $0.30 | $0.006 | $1.20 | $0.60 | Sourcechecked 2026-10-25 |
| DeepSeek V4.1 Flash 1017 | Together AI | $0.30 | $0.006 | $1.20 | none listed | Sourcechecked 2026-10-25 |
| deepseek-flashpeak hours | DeepSeek | $0.30 | $0.006 | $1.20 | none listed | Sourcechecked 2026-10-25 |
| Gemini 2.5 Flashtext, image and video input | $0.30 | $0.03 | $2.50 | $1.25 | Sourcechecked 2026-10-25 | |
| Gemini 3.5 Flash-Lite | $0.30 | $0.03 | $2.50 | $1.25 | Sourcechecked 2026-10-25 | |
| Gemini 3.5 Flash-LiteGlobal endpoint | Google Cloud | $0.30 | $0.03 | $2.50 | none listed | Sourcechecked 2026-10-25 |
| MiniMax M2.5us-east-1, in-region | Amazon Web Services | $0.30 | none listed | $1.20 | none listed | Sourcechecked 2026-10-25 |
| MiniMax M3 | Fireworks AI | $0.30 | $0.06 | $1.20 | $0.60 | Sourcechecked 2026-10-25 |
| MiniMax M3 | Together AI | $0.30 | $0.06 | $1.20 | none listed | Sourcechecked 2026-10-25 |
| MiniMax-M3prompts up to 512k tokens | MiniMax | $0.30 | $0.06 | $1.20 | none listed | Sourcechecked 2026-10-25 |
| qwen3-coder-flashprompts up to 32K | Alibaba Cloud | $0.30 | none listed | $1.50 | none listed | Sourcechecked 2026-10-25 |
| Amazon Nova 2 Liteus-east-1, in-region | Amazon Web Services | $0.33 | $0.0825 | $2.75 | none listed | Sourcechecked 2026-10-25 |
| Llama 4 Maverick | Google Cloud | $0.35 | none listed | $1.15 | $0.575 | Sourcechecked 2026-10-25 |
| qwen-plusprompts up to 256K | Alibaba Cloud | $0.40 | none listed | $1.20 | none listed | Sourcechecked 2026-10-25 |
| qwen3.7-plusprompts up to 256K | Alibaba Cloud | $0.40 | none listed | $1.60 | none listed | Sourcechecked 2026-10-25 |
| Aya Expanse (8B and 32B) | Cohere | $0.50 | none listed | $1.50 | none listed | Sourcechecked 2026-10-25 |
| Claude Haiku 5.5prompts over 100,000 tokens | Anthropic | $0.50 | $0.05 | $2.50 | $1.25 | Sourcechecked 2026-10-25 |
| Command R 03-2024 | Cohere | $0.50 | none listed | $1.50 | none listed | Sourcechecked 2026-10-25 |
| Mistral Large 3us-east-1, in-region | Amazon Web Services | $0.50 | none listed | $1.50 | none listed | Sourcechecked 2026-10-25 |
| Mistral Large 3 | Mistral AI | $0.50 | none listed | $1.50 | $0.75 | Sourcechecked 2026-10-25 |
| qwen3.6-plusprompts up to 256K | Alibaba Cloud | $0.50 | none listed | $3.00 | none listed | Sourcechecked 2026-10-25 |
| qwen3.8-27b (open source) | Alibaba Cloud | $0.50 | none listed | $3.00 | none listed | Sourcechecked 2026-10-25 |
| DeepSeek-V3.2 | Google Cloud | $0.56 | $0.056 | $1.68 | $0.84 | Sourcechecked 2026-10-25 |
| MiniMax-M3prompts over 512k tokens | MiniMax | $0.60 | $0.12 | $2.40 | none listed | Sourcechecked 2026-10-25 |
| DeepSeek v3.2us-east-1, in-region | Amazon Web Services | $0.62 | none listed | $1.85 | none listed | Sourcechecked 2026-10-25 |
| deepseek-ai/deepseek-v3.1 | Replicate | $0.672 | none listed | $2.016 | none listed | Sourcechecked 2026-10-25 |
| moonshotai/kimi-k2.5 | Replicate | $0.72 | none listed | $3.60 | none listed | Sourcechecked 2026-10-25 |
| Gemini 3.6 Flash | $0.75 | $0.075 | $3.75 | $1.875 | Sourcechecked 2026-10-25 | |
| Gemini 3.8 Flash | $0.75 | $0.075 | $3.75 | $1.875 | Sourcechecked 2026-10-25 | |
| Gemini 3.8 FlashGlobal endpoint | Google Cloud | $0.75 | $0.075 | $3.75 | none listed | Sourcechecked 2026-10-25 |
| GPT-5.4 miniGlobal | Microsoft Azure | $0.75 | $0.075 | $4.50 | $2.25 | Sourcechecked 2026-10-25 |
| gpt-5.4-mini | OpenAI | $0.75 | $0.075 | $4.50 | $2.25 | Sourcechecked 2026-10-25 |
| Qwen/Qwen3.8-27BPreview | Groq | $0.80 | none listed | $4.00 | $2.00 | Sourcechecked 2026-10-25 |
| Claude Haiku 4.5 | Anthropic | $1.00 | $0.10 | $5.00 | $2.50 | Sourcechecked 2026-10-25 |
| grok-build-0.1prompts under 200k | xAI | $1.00 | $0.20 | $2.00 | none listed | Sourcechecked 2026-10-25 |
| openai/gpt-5.6-luna | Replicate | $1.00 | none listed | $6.00 | none listed | Sourcechecked 2026-10-25 |
| Gemini 2.5 Proprompts up to 200k tokens | $1.25 | $0.125 | $10.00 | $5.00 | Sourcechecked 2026-10-25 | |
| grok-4.20-0309-non-reasoningprompts under 200k | xAI | $1.25 | $0.20 | $2.50 | $2.00 | Sourcechecked 2026-10-25 |
| grok-4.20-0309-reasoningprompts under 200k | xAI | $1.25 | $0.20 | $2.50 | $2.00 | Sourcechecked 2026-10-25 |
| grok-4.3prompts under 200k | xAI | $1.25 | $0.20 | $2.50 | $2.00 | Sourcechecked 2026-10-25 |
| Muse Spark Open | Fireworks AI | $1.25 | $0.15 | $4.25 | $2.125 | Sourcechecked 2026-10-25 |
| Muse Spark Open | Together AI | $1.25 | $0.15 | $4.25 | none listed | Sourcechecked 2026-10-25 |
| DeepSeek V4 Pro 0813 | Together AI | $1.32 | $0.13 | $3.96 | none listed | Sourcechecked 2026-10-25 |
| deepseek-v4-propeak hours | DeepSeek | $1.32 | $0.044 | $3.96 | none listed | Sourcechecked 2026-10-25 |
| perplexity/deepseek-v4-pro-0813 | Perplexity | $1.32 | $0.044 | $3.96 | none listed | Sourcechecked 2026-10-25 |
| Mistral Large 4 | Mistral AI | $1.36 | $0.14 | $4.18 | $2.09 | Sourcechecked 2026-10-25 |
| GLM 5.3 | Fireworks AI | $1.40 | $0.26 | $4.40 | $2.20 | Sourcechecked 2026-10-25 |
| GLM-5.3 | Together AI | $1.40 | $0.26 | $4.40 | none listed | Sourcechecked 2026-10-25 |
| perplexity/glm-5.3 | Perplexity | $1.40 | $0.26 | $4.40 | none listed | Sourcechecked 2026-10-25 |
| Z.ai GLM 5.3 | Mistral AI | $1.40 | $0.14 | $4.40 | $2.20 | Sourcechecked 2026-10-25 |
| Gemini 3.5 FlashGlobal endpoint | Google Cloud | $1.50 | $0.15 | $9.00 | $4.50 | Sourcechecked 2026-10-25 |
| Mistral Medium 3.5 | Mistral AI | $1.50 | none listed | $7.50 | $3.75 | Sourcechecked 2026-10-25 |
| Anthropic: Claude Sonnet 5.5 | OpenRouter | $2.00 | $0.10 | $10.00 | none listed | Sourcechecked 2026-10-25 |
| anthropic/claude-sonnet-5 | Replicate | $2.00 | none listed | $10.00 | none listed | Sourcechecked 2026-10-25 |
| Claude Sonnet 5 | Anthropic | $2.00 | $0.20 | $10.00 | $5.00 | Sourcechecked 2026-10-25 |
| Claude Sonnet 5.5 | Anthropic | $2.00 | $0.10 | $10.00 | $5.00 | Sourcechecked 2026-10-25 |
| Claude Sonnet 5.5Global endpoint | Google Cloud | $2.00 | $0.10 | $10.00 | $5.00 | Sourcechecked 2026-10-25 |
| Gemini 3.1 Pro Previewprompts up to 200k tokens | $2.00 | $0.20 | $12.00 | $6.00 | Sourcechecked 2026-10-25 | |
| Gemini 3.1 Pro Previewprompts up to 200K | Google Cloud | $2.00 | $0.20 | $12.00 | $6.00 | Sourcechecked 2026-10-25 |
| Gemini 4 Argon | $2.00 | $0.10 | $10.00 | $5.00 | Sourcechecked 2026-10-25 | |
| Gemini 4 ArgonGlobal endpoint | Google Cloud | $2.00 | $0.10 | $10.00 | $5.00 | Sourcechecked 2026-10-25 |
| Google: Gemini 4 Argon | OpenRouter | $2.00 | $0.10 | $10.00 | none listed | Sourcechecked 2026-10-25 |
| google/gemini-3.1-pro | Replicate | $2.00 | none listed | $12.00 | none listed | Sourcechecked 2026-10-25 |
| gpt-5.6-terra | OpenAI | $2.00 | $0.20 | $12.00 | $6.00 | Sourcechecked 2026-10-25 |
| GPT-5.6-terraGlobal, short context | Microsoft Azure | $2.00 | $0.20 | $12.00 | none listed | Sourcechecked 2026-10-25 |
| gpt-6-sol | OpenAI | $2.00 | $0.20 | $10.00 | $5.00 | Sourcechecked 2026-10-25 |
| GPT-6.1 SolGlobal, short context | Microsoft Azure | $2.00 | $0.10 | $10.00 | none listed | Sourcechecked 2026-10-25 |
| gpt-6.1-sol | OpenAI | $2.00 | $0.10 | $10.00 | $5.00 | Sourcechecked 2026-10-25 |
| grok-4.5prompts under 200k | xAI | $2.00 | $0.30 | $6.00 | $4.80 | Sourcechecked 2026-10-25 |
| grok-4.6prompts under 200k | xAI | $2.00 | $0.50 | $6.00 | $4.80 | Sourcechecked 2026-10-25 |
| grok-4.7prompts under 200k | xAI | $2.00 | $0.50 | $6.00 | $4.80 | Sourcechecked 2026-10-25 |
| MoonshotAI: Kimi K3.1 Pro | OpenRouter | $2.00 | $0.20 | $10.00 | none listed | Sourcechecked 2026-10-25 |
| OpenAI: GPT-6.1 Sol | OpenRouter | $2.00 | $0.10 | $10.00 | none listed | Sourcechecked 2026-10-25 |
| openai/gpt-5.6-sol | Replicate | $2.00 | none listed | $10.00 | none listed | Sourcechecked 2026-10-25 |
| Qwen 3.8 Max | Fireworks AI | $2.00 | $0.25 | $6.00 | $3.00 | Sourcechecked 2026-10-25 |
| Qwen3.8-2.4T-A95B | Together AI | $2.00 | $0.25 | $6.00 | none listed | Sourcechecked 2026-10-25 |
| qwen3.8-max | Alibaba Cloud | $2.00 | none listed | $6.00 | none listed | Sourcechecked 2026-10-25 |
| SpaceXAI: Grok 4.7 | OpenRouter | $2.00 | $0.50 | $6.00 | none listed | Sourcechecked 2026-10-25 |
| Claude Sonnet 5.5us-east-1, in-region | Amazon Web Services | $2.20 | $0.11 | $11.00 | none listed | Sourcechecked 2026-10-25 |
| Command R+ 08-2024 | Cohere | $2.50 | none listed | $10.00 | none listed | Sourcechecked 2026-10-25 |
| Gemini 2.5 Proprompts over 200k tokens | $2.50 | $0.25 | $15.00 | $7.50 | Sourcechecked 2026-10-25 | |
| GPT-5.4Global, under 272k context | Microsoft Azure | $2.50 | $0.25 | $15.00 | $7.50 | Sourcechecked 2026-10-25 |
| qwen3.7-max | Alibaba Cloud | $2.50 | none listed | $7.50 | none listed | Sourcechecked 2026-10-25 |
| Claude Sonnet 4.6 | Anthropic | $3.00 | $0.30 | $15.00 | $7.50 | Sourcechecked 2026-10-25 |
| Ember-1 | Fireworks AI | $3.00 | $0.30 | $15.00 | $7.50 | Sourcechecked 2026-10-25 |
| Kimi K3 | Fireworks AI | $3.00 | $0.30 | $15.00 | $7.50 | Sourcechecked 2026-10-25 |
| Kimi K3 | Together AI | $3.00 | $0.30 | $15.00 | none listed | Sourcechecked 2026-10-25 |
| perplexity/kimi-k3 | Perplexity | $3.00 | $0.30 | $15.00 | none listed | Sourcechecked 2026-10-25 |
| deepseek-ai/deepseek-r1 | Replicate | $3.75 | none listed | $10.00 | none listed | Sourcechecked 2026-10-25 |
| Anthropic: Claude Opus 5.5 | OpenRouter | $4.00 | $0.20 | $20.00 | none listed | Sourcechecked 2026-10-25 |
| anthropic/claude-opus-5.5 | Replicate | $4.00 | none listed | $20.00 | none listed | Sourcechecked 2026-10-25 |
| Claude Opus 5.5 | Anthropic | $4.00 | $0.20 | $20.00 | $10.00 | Sourcechecked 2026-10-25 |
| Claude Opus 5.5Global endpoint | Google Cloud | $4.00 | $0.20 | $20.00 | none listed | Sourcechecked 2026-10-25 |
| Gemini 3.1 Pro Previewprompts over 200k tokens | $4.00 | $0.40 | $18.00 | $9.00 | Sourcechecked 2026-10-25 | |
| gpt-5.6-sol | OpenAI | $4.00 | $0.40 | $20.00 | $10.00 | Sourcechecked 2026-10-25 |
| GPT-5.6-solGlobal, short context | Microsoft Azure | $4.00 | $0.40 | $20.00 | none listed | Sourcechecked 2026-10-25 |
| grok-4.7prompts 200k and over | xAI | $4.00 | $1.00 | $12.00 | $9.60 | Sourcechecked 2026-10-25 |
| Claude Opus 5.5us-east-1, in-region | Amazon Web Services | $4.40 | $0.22 | $22.00 | none listed | Sourcechecked 2026-10-25 |
| Claude Opus 5 | Anthropic | $5.00 | $0.50 | $25.00 | $12.50 | Sourcechecked 2026-10-25 |
| gpt-5.5prompts up to 272K | OpenAI | $5.00 | $0.50 | $30.00 | $15.00 | Sourcechecked 2026-10-25 |
| GPT-5.5Global | Microsoft Azure | $5.00 | $0.50 | $30.00 | none listed | Sourcechecked 2026-10-25 |
| anthropic/claude-fable-5 | Replicate | $10.00 | none listed | $50.00 | none listed | Sourcechecked 2026-10-25 |
| Claude Fable 5 | Anthropic | $10.00 | $1.00 | $50.00 | $25.00 | Sourcechecked 2026-10-25 |
| Claude Fable 5.1 | Anthropic | $10.00 | $0.25 | $50.00 | $25.00 | Sourcechecked 2026-10-25 |
| GPT-6 AstraGlobal, short context | Microsoft Azure | $10.00 | $1.00 | $50.00 | none listed | Sourcechecked 2026-10-25 |
| gpt-6-astra | OpenAI | $10.00 | $1.00 | $50.00 | $25.00 | Sourcechecked 2026-10-25 |
Terms on this page: Token · Prompt caching · Context window · Batch API
Estimator
Estimate one workload across every vendor
Enter your monthly tokens and their split. The most expensive model here charges $50.00 per 1M output tokens, so the spread is wide.
Write 40k, 250M or 5B. Input, cached input and output tokens together.
100M tokens a month costs from $2.50 (Amazon Nova Micro) to $1,200 (anthropic/claude-fable-5).
| Model | Vendor | Per month | Per 1M tokens at this split |
|---|---|---|---|
| Amazon Nova Microus-east-1, in-region | Amazon Web Services | $2.50 | $0.025025 |
| NVIDIA Nemotron 3.5 Lightning 30B A3B | Fireworks AI | $3.43 | $0.0343 |
| qwen3.7-flashprompts up to 32K | Alibaba Cloud | $3.50 | $0.035 |
| Command R7B | Cohere | $4.31 | $0.043125 |
| Gemini 2.5 Flash-Litetext, image and video input | $6.28 | $0.0628 | |
| GPT OSS 20B | Groq | $6.42 | $0.06421 |
| qwen-flashprompts up to 256K | Alibaba Cloud | $6.75 | $0.0675 |
| gpt-6-luna | OpenAI | $6.78 | $0.0678 |
| Claude Haiku 5.5prompts up to 100,000 tokens | Anthropic | $6.78 | $0.0678 |
| Claude Haiku 5.5prompts up to 200K | Google Cloud | $6.78 | $0.0678 |
| Anthropic: Claude Haiku 5.5 | OpenRouter | $6.78 | $0.0678 |
| OpenAI: GPT-6 Luna | OpenRouter | $6.78 | $0.0678 |
| Claude Haiku 5.5us-east-1, in-region | Amazon Web Services | $7.46 | $0.07458 |
| Safety GPT OSS 20BPreview | Groq | $8.63 | $0.08625 |
| DeepSeek: DeepSeek V4.1 Flash 1017 | OpenRouter | $8.72 | $0.08724 |
| MiniMax-M3.1-Flash | MiniMax | $9.42 | $0.0942 |
| OpenAI GPT OSS 120B | Fireworks AI | $9.42 | $0.0942 |
| GLM 5.3 Flash | Fireworks AI | $9.79 | $0.0979 |
| perplexity/glm-5.3-flash | Perplexity | $9.79 | $0.0979 |
| GLM-5.3-Flash | Together AI | $9.79 | $0.0979 |
| Ministral 3 3B | Mistral AI | $10.00 | $0.10 |
| perplexity/nemotron-3.5-super-120b-a12b | Perplexity | $12.22 | $0.1222 |
| GPT OSS 120B | Groq | $12.90 | $0.129 |
| OpenAI: gpt-oss-120b | OpenRouter | $12.90 | $0.129 |
| gpt-5.6-luna | OpenAI | $14.56 | $0.1456 |
| GPT-5.6-lunaGlobal, short context | Microsoft Azure | $14.56 | $0.1456 |
| gpt-5.4-nano | OpenAI | $14.81 | $0.1481 |
| GPT-5.4 nanoGlobal | Microsoft Azure | $14.81 | $0.1481 |
| Ministral 3 8B | Mistral AI | $15.00 | $0.15 |
| qwen3.8-flash | Alibaba Cloud | $16.60 | $0.166 |
| gpt-oss-120bus-east-1, in-region | Amazon Web Services | $17.25 | $0.1725 |
| Command R | Cohere | $17.25 | $0.1725 |
| Mistral Small 4 | Mistral AI | $17.25 | $0.1725 |
| gpt-oss-120B | Together AI | $17.25 | $0.1725 |
| Qwen3.5 9B | Together AI | $17.40 | $0.174 |
| deepseek-flashpeak hours | DeepSeek | $17.45 | $0.17448 |
| DeepSeek V4.1 Flash | Fireworks AI | $17.45 | $0.17448 |
| DeepSeek V4.1 Flash 1017 | Together AI | $17.45 | $0.17448 |
| Gemini 3.1 Flash-Litetext, image and video input | $18.20 | $0.182 | |
| Ministral 3 14B | Mistral AI | $20.00 | $0.20 |
| MiniMax-M3prompts up to 512k tokens | MiniMax | $20.58 | $0.2058 |
| MiniMax M3 | Fireworks AI | $20.58 | $0.2058 |
| MiniMax M3 | Together AI | $20.58 | $0.2058 |
| openai/gpt-oss-120b | Replicate | $20.70 | $0.207 |
| Gemini 3.5 Flash-Lite | $25.34 | $0.2534 | |
| Gemini 2.5 Flashtext, image and video input | $25.34 | $0.2534 | |
| Gemini 3.5 Flash-LiteGlobal endpoint | Google Cloud | $25.34 | $0.2534 |
| perplexity/sonar | Perplexity | $25.38 | $0.25375 |
| Llama 4 Maverick 17Bus-east-1, in-region | Amazon Web Services | $27.65 | $0.2765 |
| Amazon Nova 2 Liteus-east-1, in-region | Amazon Web Services | $30.75 | $0.30745 |
| DeepSeek-V3.2 | Google Cloud | $32.37 | $0.32368 |
| Codestral | Mistral AI | $33.00 | $0.33 |
| MiniMax M2.5us-east-1, in-region | Amazon Web Services | $34.50 | $0.345 |
| qwen3-coder-flashprompts up to 32K | Alibaba Cloud | $36.00 | $0.36 |
| Llama 4 Maverick | Google Cloud | $39.00 | $0.39 |
| qwen-plusprompts up to 256K | Alibaba Cloud | $44.00 | $0.44 |
| qwen3.7-plusprompts up to 256K | Alibaba Cloud | $46.00 | $0.46 |
| Gemini 3.8 Flash | $50.85 | $0.5085 | |
| Gemini 3.6 Flash | $50.85 | $0.5085 | |
| Gemini 3.8 FlashGlobal endpoint | Google Cloud | $50.85 | $0.5085 |
| gpt-5.4-mini | OpenAI | $54.60 | $0.546 |
| GPT-5.4 miniGlobal | Microsoft Azure | $54.60 | $0.546 |
| Mistral Large 3us-east-1, in-region | Amazon Web Services | $55.00 | $0.55 |
| Aya Expanse (8B and 32B) | Cohere | $55.00 | $0.55 |
| Command R 03-2024 | Cohere | $55.00 | $0.55 |
| Mistral Large 3 | Mistral AI | $55.00 | $0.55 |
| grok-build-0.1prompts under 200k | xAI | $58.60 | $0.586 |
| qwen3.6-plusprompts up to 256K | Alibaba Cloud | $62.50 | $0.625 |
| qwen3.8-27b (open source) | Alibaba Cloud | $62.50 | $0.625 |
| Claude Haiku 4.5 | Anthropic | $67.80 | $0.678 |
| DeepSeek v3.2us-east-1, in-region | Amazon Web Services | $68.15 | $0.6815 |
| grok-4.3prompts under 200k | xAI | $70.35 | $0.7035 |
| grok-4.20-0309-reasoningprompts under 200k | xAI | $70.35 | $0.7035 |
| grok-4.20-0309-non-reasoningprompts under 200k | xAI | $70.35 | $0.7035 |
| deepseek-v4-propeak hours | DeepSeek | $71.19 | $0.71192 |
| perplexity/deepseek-v4-pro-0813 | Perplexity | $71.19 | $0.71192 |
| deepseek-ai/deepseek-v3.1 | Replicate | $73.92 | $0.7392 |
| DeepSeek V4 Pro 0813 | Together AI | $76.18 | $0.7618 |
| Muse Spark Open | Fireworks AI | $76.20 | $0.762 |
| Muse Spark Open | Together AI | $76.20 | $0.762 |
| Mistral Large 4 | Mistral AI | $79.34 | $0.7934 |
| Z.ai GLM 5.3 | Mistral AI | $81.92 | $0.8192 |
| moonshotai/kimi-k2.5 | Replicate | $86.40 | $0.864 |
| GLM 5.3 | Fireworks AI | $88.88 | $0.8888 |
| perplexity/glm-5.3 | Perplexity | $88.88 | $0.8888 |
| GLM-5.3 | Together AI | $88.88 | $0.8888 |
| Qwen/Qwen3.8-27BPreview | Groq | $96.00 | $0.96 |
| Gemini 2.5 Proprompts up to 200k tokens | $103.50 | $1.035 | |
| Gemini 3.5 FlashGlobal endpoint | Google Cloud | $109.20 | $1.092 |
| Qwen 3.8 Max | Fireworks AI | $118.50 | $1.185 |
| Qwen3.8-2.4T-A95B | Together AI | $118.50 | $1.185 |
| grok-4.5prompts under 200k | xAI | $121.40 | $1.214 |
| openai/gpt-5.6-luna | Replicate | $125.00 | $1.25 |
| gpt-6.1-sol | OpenAI | $129.80 | $1.298 |
| Claude Sonnet 5.5 | Anthropic | $129.80 | $1.298 |
| Gemini 4 Argon | $129.80 | $1.298 | |
| GPT-6.1 SolGlobal, short context | Microsoft Azure | $129.80 | $1.298 |
| Gemini 4 ArgonGlobal endpoint | Google Cloud | $129.80 | $1.298 |
| Claude Sonnet 5.5Global endpoint | Google Cloud | $129.80 | $1.298 |
| Anthropic: Claude Sonnet 5.5 | OpenRouter | $129.80 | $1.298 |
| OpenAI: GPT-6.1 Sol | OpenRouter | $129.80 | $1.298 |
| Google: Gemini 4 Argon | OpenRouter | $129.80 | $1.298 |
| SpaceXAI: Grok 4.7 | OpenRouter | $133.00 | $1.33 |
| grok-4.7prompts under 200k | xAI | $133.00 | $1.33 |
| grok-4.6prompts under 200k | xAI | $133.00 | $1.33 |
| gpt-6-sol | OpenAI | $135.60 | $1.356 |
| Claude Sonnet 5 | Anthropic | $135.60 | $1.356 |
| MoonshotAI: Kimi K3.1 Pro | OpenRouter | $135.60 | $1.356 |
| Claude Sonnet 5.5us-east-1, in-region | Amazon Web Services | $142.78 | $1.4278 |
| gpt-5.6-terra | OpenAI | $145.60 | $1.456 |
| Gemini 3.1 Pro Previewprompts up to 200k tokens | $145.60 | $1.456 | |
| GPT-5.6-terraGlobal, short context | Microsoft Azure | $145.60 | $1.456 |
| Gemini 3.1 Pro Previewprompts up to 200K | Google Cloud | $145.60 | $1.456 |
| Mistral Medium 3.5 | Mistral AI | $180.00 | $1.80 |
| GPT-5.4Global, under 272k context | Microsoft Azure | $182.00 | $1.82 |
| Claude Sonnet 4.6 | Anthropic | $203.40 | $2.034 |
| Ember-1 | Fireworks AI | $203.40 | $2.034 |
| Kimi K3 | Fireworks AI | $203.40 | $2.034 |
| perplexity/kimi-k3 | Perplexity | $203.40 | $2.034 |
| Kimi K3 | Together AI | $203.40 | $2.034 |
| qwen3.8-max | Alibaba Cloud | $220.00 | $2.20 |
| anthropic/claude-sonnet-5 | Replicate | $240.00 | $2.40 |
| openai/gpt-5.6-sol | Replicate | $240.00 | $2.40 |
| google/gemini-3.1-pro | Replicate | $250.00 | $2.50 |
| Claude Opus 5.5 | Anthropic | $259.60 | $2.596 |
| Claude Opus 5.5Global endpoint | Google Cloud | $259.60 | $2.596 |
| Anthropic: Claude Opus 5.5 | OpenRouter | $259.60 | $2.596 |
| gpt-5.6-sol | OpenAI | $271.20 | $2.712 |
| GPT-5.6-solGlobal, short context | Microsoft Azure | $271.20 | $2.712 |
| qwen3.7-max | Alibaba Cloud | $275.00 | $2.75 |
| Claude Opus 5.5us-east-1, in-region | Amazon Web Services | $285.56 | $2.8556 |
| Command R+ 08-2024 | Cohere | $287.50 | $2.875 |
| Claude Opus 5 | Anthropic | $339.00 | $3.39 |
| gpt-5.5prompts up to 272K | OpenAI | $364.00 | $3.64 |
| GPT-5.5Global | Microsoft Azure | $364.00 | $3.64 |
| deepseek-ai/deepseek-r1 | Replicate | $406.25 | $4.0625 |
| anthropic/claude-opus-5.5 | Replicate | $480.00 | $4.80 |
| Claude Fable 5.1 | Anthropic | $634.50 | $6.345 |
| gpt-6-astra | OpenAI | $678.00 | $6.78 |
| Claude Fable 5 | Anthropic | $678.00 | $6.78 |
| GPT-6 AstraGlobal, short context | Microsoft Azure | $678.00 | $6.78 |
| anthropic/claude-fable-5 | Replicate | $1,200 | $12.00 |
Standard prices in USD on 2026-10-25, before tax. Long-prompt tiers are left out; see the price table.
How to choose
How to choose an LLM API on price
Price your split, not the headline
Measure how many of your tokens are new input, cached input and output, then compare at that split. The cheapest input price is often not the cheapest bill.
Cache what repeats
Long system prompts, tool definitions and reference documents sent on every call are what prompt caching is for. Where a vendor lists a cached price, it runs from half to a fiftieth of the input price.
Batch what can wait
Overnight classification, back-filling and evaluation runs rarely need an answer in seconds. Batch tiers halve the bill where they exist.
Read the tiers and the dates
Some models charge more for long prompts, and some prices are promotions with an end date. Both are in the table; tick "After promotions end" to price the later rates.
Test on your own prompts
Tokenizers differ, so the same text is a different number of tokens on each model. Anthropic, for one, says its newer tokenizer produces about 30% more tokens for the same text.
Keep a fallback
Prices and models change often. Routing through one interface lets you move work to another model when the price or quality shifts.
FAQ
LLM API pricing questions
Which LLM API is cheapest?
By list price per input token, qwen3.7-flash at $0.03 per 1M tokens, checked 2026-10-25. The cheapest model for your app depends on how your tokens split between new input, cached input and output, so run your own numbers in the estimator.
Why does the token split matter so much?
Output tokens usually cost several times more than input tokens, and cached input costs a fraction of input. Two apps with the same monthly tokens can get very different bills: a chat app that writes long answers pays mostly for output, a document reader that re-sends the same instructions pays mostly for cached input.
How do batch and off-peak prices compare?
OpenAI, Anthropic, Google, Fireworks AI, Groq and Mistral take 50% off for batch work that can wait, and xAI takes 20% off. Some vendors list a batch price only on some models; the table says "none listed" where there is none. DeepSeek has no batch tier but halves its prices off-peak.
How current are these prices?
Every row was read off the vendor's own pricing page on 2026-10-25 and links to it. A price we cannot confirm shows "check the vendor's page" instead of a number, and a promotional price disappears once its end date passes.
Related
Vendor pages and tools
OpenAI API pricing10 models and 20 other prices, checked 2026-10-25.Claude API pricing9 models, checked 2026-10-25.Gemini API pricing9 models and 9 other prices, checked 2026-10-25.DeepSeek API pricing2 models, checked 2026-10-25.MiniMax API pricing2 models, checked 2026-10-25.Qwen API pricing10 models and 2 other prices, checked 2026-10-25.Amazon Bedrock pricing10 models and 7 other prices, checked 2026-10-25.Azure OpenAI pricing9 models and 7 other prices, checked 2026-10-25.Cohere API pricing5 models and 2 other prices, checked 2026-10-25.Fireworks AI API pricing10 models and 3 other prices, checked 2026-10-25.Vertex AI pricing10 models and 4 other prices, checked 2026-10-25.Groq API pricing4 models and 4 other prices, checked 2026-10-25.Mistral API pricing9 models and 6 other prices, checked 2026-10-25.OpenRouter API pricing10 models, checked 2026-10-25.Perplexity API pricing6 models and 4 other prices, checked 2026-10-25.Replicate API pricing10 models and 12 other prices, checked 2026-10-25.Together AI API pricing10 models and 11 other prices, checked 2026-10-25.Grok API pricing7 models and 6 other prices, checked 2026-10-25.AssemblyAI API pricing7 prices, checked 2026-10-25.Deepgram API pricing11 prices, checked 2026-10-25.ElevenLabs API pricing9 prices, checked 2026-10-25.LLM API cost calculatorPrice one workload across every vendor and model at once.Prompt caching savingsSee what caching a long system prompt saves each month.All pricing pagesEvery vendor we track, with the day each was checked.
Choosing a model for a real workflow? OceanLab builds AI into business workflows, with people kept in charge. AI automation →
