LLM Token Cost Calculator
Know exactly what your AI will cost before you build it. Compare real-time per-1M-token pricing across GPT, Claude, Gemini, and more - then estimate your real monthly spend on input and output tokens in seconds.
Dial in your token volume
1,000,000
1,000,000
Live Cost Comparison
Anthropic
Claude Opus 4.1Flagship
$90.00
/ mo
Claude Opus 4Flagship
$90.00
/ mo
Claude Fable 5.1Flagship
$60.00
/ mo
Claude Fable 5Flagship
$60.00
/ mo
DeepSeek
DeepSeek V4 Pro 0423Mid
$2.87
/ mo
R1 Distill Llama 70BMid
$1.60
/ mo
R1Mid
$3.20
/ mo
DeepSeek V4 Pro 0813Mid
$2.32
/ mo
Google
Nano Banana Pro (Gemini 3 Pro Image)Mid
$14.00
/ mo
Gemini 3.5 FlashMid
$10.50
/ mo
Gemini 2.5 ProMid
$11.25
/ mo
Gemini 3.8 FlashMid
$4.50
/ mo
OpenAI
o1-proFlagship
$750.00
/ mo
GPT-5.5 ProFlagship
$210.00
/ mo
GPT-5.4 ProFlagship
$210.00
/ mo
GPT-4Flagship
$90.00
/ mo
Qwen
Qwen3.8 Max (0902)Mid
$8.00
/ mo
Qwen3.8 2.4T A95BMid
$8.00
/ mo
Qwen3.7 MaxMid
$5.90
/ mo
Qwen3 Max ThinkingMid
$4.68
/ mo
SpaceXAI
Grok 4.6Mid
$8.00
/ mo
Grok 4.5Mid
$8.00
/ mo
Grok 4.3Mid
$3.75
/ mo
Grok 4.20 Multi-AgentMid
$3.75
/ mo
Full pricing comparison (per 1M tokens)
| Provider | Model | Tier | Input $/1M | Output $/1M | Blended* $/1M |
|---|---|---|---|---|---|
| Anthropic | Claude Opus 4.1 | Flagship | $15.00 | $75.00 | $30.00 |
| Anthropic | Claude Opus 4 | Flagship | $15.00 | $75.00 | $30.00 |
| Anthropic | Claude Fable 5.1 | Flagship | $10.00 | $50.00 | $20.00 |
| Anthropic | Claude Fable 5 | Flagship | $10.00 | $50.00 | $20.00 |
| Anthropic | Claude Opus 5 | Flagship | $5.00 | $25.00 | $10.00 |
| Anthropic | Claude Opus 4.8 | Flagship | $5.00 | $25.00 | $10.00 |
| Anthropic | Claude Opus 4.7 | Flagship | $5.00 | $25.00 | $10.00 |
| Anthropic | Claude Opus 4.6 | Flagship | $5.00 | $25.00 | $10.00 |
| Anthropic | Claude Opus 4.5 | Flagship | $5.00 | $25.00 | $10.00 |
| Anthropic | Claude Sonnet 4.6 | Flagship | $3.00 | $15.00 | $6.00 |
| Anthropic | Claude Sonnet 4.5 | Flagship | $3.00 | $15.00 | $6.00 |
| Anthropic | Claude Sonnet 4 | Flagship | $3.00 | $15.00 | $6.00 |
| Anthropic | Claude Sonnet 5 | Mid | $2.00 | $10.00 | $4.00 |
| Anthropic | Claude Haiku 4.5 | Mid | $1.00 | $5.00 | $2.00 |
| Anthropic | Claude 3 Haiku | Budget | $0.25 | $1.25 | $0.50 |
| DeepSeek | DeepSeek V4 Pro 0423 | Mid | $0.96 | $1.91 | $1.19 |
| DeepSeek | R1 Distill Llama 70B | Mid | $0.80 | $0.80 | $0.80 |
| DeepSeek | R1 | Mid | $0.70 | $2.50 | $1.15 |
| DeepSeek | DeepSeek V4 Pro 0813 | Mid | $0.58 | $1.74 | $0.87 |
| DeepSeek | R1 0528 | Budget | $0.50 | $2.15 | $0.91 |
| DeepSeek | DeepSeek V3 | Budget | $0.32 | $0.89 | $0.46 |
| DeepSeek | DeepSeek V3 0324 | Budget | $0.29 | $1.14 | $0.50 |
| DeepSeek | DeepSeek V3.2 Exp | Budget | $0.27 | $0.41 | $0.30 |
| DeepSeek | DeepSeek V3.1 Terminus | Budget | $0.27 | $1.00 | $0.45 |
| DeepSeek | DeepSeek V3.2 | Budget | $0.27 | $0.40 | $0.30 |
| DeepSeek | DeepSeek V3.1 | Budget | $0.25 | $0.95 | $0.42 |
| DeepSeek | DeepSeek V4 Flash 0423 | Budget | $0.09 | $0.18 | $0.11 |
| DeepSeek | DeepSeek V4 Flash 0731 | Budget | $0.07 | $0.18 | $0.09 |
| Nano Banana Pro (Gemini 3 Pro Image) | Mid | $2.00 | $12.00 | $4.50 | |
| Gemini 3.5 Flash | Mid | $1.50 | $9.00 | $3.38 | |
| Gemini 2.5 Pro | Mid | $1.25 | $10.00 | $3.44 | |
| Gemini 3.8 Flash | Mid | $0.75 | $3.75 | $1.50 | |
| Gemini 3.7 Flash | Mid | $0.75 | $3.75 | $1.50 | |
| Gemini 3.6 Flash | Mid | $0.75 | $3.75 | $1.50 | |
| Gemma 2 27B | Mid | $0.65 | $0.65 | $0.65 | |
| Nano Banana 2 (Gemini 3.1 Flash Image) | Budget | $0.50 | $3.00 | $1.13 | |
| Gemini 3.5 Flash Lite | Budget | $0.30 | $2.50 | $0.85 | |
| Nano Banana (Gemini 2.5 Flash Image) | Budget | $0.30 | $2.50 | $0.85 | |
| Gemini 2.5 Flash | Budget | $0.30 | $2.50 | $0.85 | |
| Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | Budget | $0.25 | $1.50 | $0.56 | |
| Gemini 3.1 Flash Lite | Budget | $0.25 | $1.50 | $0.56 | |
| Gemini 2.5 Flash Lite | Budget | $0.10 | $0.40 | $0.17 | |
| Gemma 4 31B | Budget | $0.09 | $0.34 | $0.15 | |
| Gemma 3 27B | Budget | $0.08 | $0.45 | $0.17 | |
| Gemma 4 26B A4B | Budget | $0.07 | $0.34 | $0.14 | |
| Gemma 3 4B | Budget | $0.05 | $0.10 | $0.06 | |
| Gemma 3 12B | Budget | $0.05 | $0.15 | $0.07 | |
| OpenAI | o1-pro | Flagship | $150.00 | $600.00 | $262.50 |
| OpenAI | GPT-5.5 Pro | Flagship | $30.00 | $180.00 | $67.50 |
| OpenAI | GPT-5.4 Pro | Flagship | $30.00 | $180.00 | $67.50 |
| OpenAI | GPT-4 | Flagship | $30.00 | $60.00 | $37.50 |
| OpenAI | GPT-5.2 Pro | Flagship | $21.00 | $168.00 | $57.75 |
| OpenAI | o3 Pro | Flagship | $20.00 | $80.00 | $35.00 |
| OpenAI | GPT-5 Pro | Flagship | $15.00 | $120.00 | $41.25 |
| OpenAI | o1 | Flagship | $15.00 | $60.00 | $26.25 |
| OpenAI | GPT-6 Astra | Flagship | $10.00 | $50.00 | $20.00 |
| OpenAI | GPT-6 Astra Pro | Flagship | $10.00 | $50.00 | $20.00 |
| OpenAI | GPT-5 Image | Flagship | $10.00 | $10.00 | $10.00 |
| OpenAI | GPT-4 Turbo | Flagship | $10.00 | $30.00 | $15.00 |
| OpenAI | GPT-5.4 Image 2 | Flagship | $8.00 | $15.00 | $9.75 |
| OpenAI | GPT-5.5 | Flagship | $5.00 | $30.00 | $11.25 |
| OpenAI | GPT-4o (2024-05-13) | Flagship | $5.00 | $15.00 | $7.50 |
| OpenAI | GPT-3.5 Turbo 16k | Flagship | $3.00 | $4.00 | $3.25 |
| OpenAI | GPT-5.4 | Mid | $2.50 | $15.00 | $5.63 |
| OpenAI | GPT Audio | Mid | $2.50 | $10.00 | $4.38 |
| OpenAI | GPT-5 Image Mini | Mid | $2.50 | $2.00 | $2.38 |
| OpenAI | GPT-4o (2024-11-20) | Mid | $2.50 | $10.00 | $4.38 |
| OpenAI | GPT-4o (2024-08-06) | Mid | $2.50 | $10.00 | $4.38 |
| OpenAI | GPT-4o | Mid | $2.50 | $10.00 | $4.38 |
| OpenAI | GPT-5.6 Terra Pro | Mid | $2.00 | $12.00 | $4.50 |
| OpenAI | GPT-5.6 Terra | Mid | $2.00 | $12.00 | $4.50 |
| OpenAI | GPT-5.6 Sol Pro | Mid | $2.00 | $10.00 | $4.00 |
| OpenAI | GPT-5.6 Sol | Mid | $2.00 | $10.00 | $4.00 |
| OpenAI | o3 | Mid | $2.00 | $8.00 | $3.50 |
| OpenAI | GPT-4.1 | Mid | $2.00 | $8.00 | $3.50 |
| OpenAI | GPT-5.3-Codex | Mid | $1.75 | $14.00 | $4.81 |
| OpenAI | GPT-5.2-Codex | Mid | $1.75 | $14.00 | $4.81 |
| OpenAI | GPT-5.2 Chat | Mid | $1.75 | $14.00 | $4.81 |
| OpenAI | GPT-5.2 | Mid | $1.75 | $14.00 | $4.81 |
| OpenAI | GPT-5.1-Codex-Max | Mid | $1.25 | $10.00 | $3.44 |
| OpenAI | GPT-5.1 | Mid | $1.25 | $10.00 | $3.44 |
| OpenAI | GPT-5.1-Codex | Mid | $1.25 | $10.00 | $3.44 |
| OpenAI | GPT-5 | Mid | $1.25 | $10.00 | $3.44 |
| OpenAI | o4 Mini High | Mid | $1.10 | $4.40 | $1.93 |
| OpenAI | o4 Mini | Mid | $1.10 | $4.40 | $1.93 |
| OpenAI | o3 Mini High | Mid | $1.10 | $4.40 | $1.93 |
| OpenAI | o3 Mini | Mid | $1.10 | $4.40 | $1.93 |
| OpenAI | GPT-3.5 Turbo (older v0613) | Mid | $1.00 | $2.00 | $1.25 |
| OpenAI | GPT-5.4 Mini | Mid | $0.75 | $4.50 | $1.69 |
| OpenAI | GPT Audio Mini | Mid | $0.60 | $2.40 | $1.05 |
| OpenAI | GPT-3.5 Turbo | Budget | $0.50 | $1.50 | $0.75 |
| OpenAI | GPT-4.1 Mini | Budget | $0.40 | $1.60 | $0.70 |
| OpenAI | GPT-5.1-Codex-Mini | Budget | $0.25 | $2.00 | $0.69 |
| OpenAI | GPT-5 Mini | Budget | $0.25 | $2.00 | $0.69 |
| OpenAI | GPT-5.6 Luna Pro | Budget | $0.20 | $1.20 | $0.45 |
| OpenAI | GPT-5.6 Luna | Budget | $0.20 | $1.20 | $0.45 |
| OpenAI | GPT-5.4 Nano | Budget | $0.20 | $1.25 | $0.46 |
| OpenAI | GPT-4o-mini | Budget | $0.15 | $0.60 | $0.26 |
| OpenAI | GPT-4o-mini (2024-07-18) | Budget | $0.15 | $0.60 | $0.26 |
| OpenAI | GPT-4.1 Nano | Budget | $0.10 | $0.40 | $0.17 |
| OpenAI | gpt-oss-safeguard-20b | Budget | $0.07 | $0.30 | $0.13 |
| OpenAI | GPT-5 Nano | Budget | $0.05 | $0.40 | $0.14 |
| OpenAI | gpt-oss-120b | Budget | $0.04 | $0.17 | $0.07 |
| OpenAI | gpt-oss-20b | Budget | $0.03 | $0.13 | $0.06 |
| Qwen | Qwen3.8 Max (0902) | Mid | $2.00 | $6.00 | $3.00 |
| Qwen | Qwen3.8 2.4T A95B | Mid | $2.00 | $6.00 | $3.00 |
| Qwen | Qwen3.7 Max | Mid | $1.48 | $4.42 | $2.21 |
| Qwen | Qwen3 Max Thinking | Mid | $0.78 | $3.90 | $1.56 |
| Qwen | Qwen3 Max | Mid | $0.78 | $3.90 | $1.56 |
| Qwen | Qwen3 Coder Plus | Mid | $0.65 | $3.25 | $1.30 |
| Qwen | Qwen3.5 397B A17B | Mid | $0.55 | $3.50 | $1.29 |
| Qwen | Qwen3 235B A22B | Budget | $0.45 | $1.82 | $0.80 |
| Qwen | Qwen3.8 27B | Budget | $0.42 | $3.00 | $1.06 |
| Qwen | Qwen3 VL 235B A22B Thinking | Budget | $0.40 | $4.00 | $1.30 |
| Qwen | Qwen3.6 Plus | Budget | $0.33 | $1.95 | $0.73 |
| Qwen | Qwen3.7 Plus | Budget | $0.32 | $1.28 | $0.56 |
| Qwen | Qwen3.5-35B-A3B | Budget | $0.31 | $1.25 | $0.55 |
| Qwen | Qwen3.5 Plus 2026-04-20 | Budget | $0.30 | $1.80 | $0.67 |
| Qwen | Qwen3.6 27B | Budget | $0.30 | $2.00 | $0.72 |
| Qwen | Qwen3 Coder 480B A35B | Budget | $0.30 | $1.00 | $0.47 |
| Qwen | Qwen3.5-122B-A10B | Budget | $0.29 | $2.40 | $0.82 |
| Qwen | Qwen3.5 Plus 2026-02-15 | Budget | $0.26 | $1.56 | $0.58 |
| Qwen | Qwen Plus 0728 | Budget | $0.26 | $0.78 | $0.39 |
| Qwen | Qwen-Plus | Budget | $0.26 | $0.78 | $0.39 |
| Qwen | Qwen3 235B A22B Thinking 2507 | Budget | $0.23 | $2.30 | $0.75 |
| Qwen | Qwen3 14B | Budget | $0.23 | $0.91 | $0.40 |
| Qwen | Qwen3 VL 30B A3B Thinking | Budget | $0.20 | $2.40 | $0.75 |
| Qwen | Qwen3 30B A3B Thinking 2507 | Budget | $0.20 | $2.40 | $0.75 |
| Qwen | Qwen3.5-27B | Budget | $0.20 | $1.56 | $0.54 |
| Qwen | Qwen3 Coder Flash | Budget | $0.20 | $0.97 | $0.39 |
| Qwen | Qwen3.6 Flash | Budget | $0.19 | $1.13 | $0.42 |
| Qwen | Qwen3 VL 8B Thinking | Budget | $0.18 | $2.10 | $0.66 |
| Qwen | Qwen3.8 Flash | Budget | $0.15 | $0.47 | $0.23 |
| Qwen | Qwen3 Next 80B A3B Thinking | Budget | $0.15 | $1.20 | $0.41 |
| Qwen | Qwen3 Coder Next | Budget | $0.12 | $0.80 | $0.29 |
| Qwen | Qwen3 30B A3B | Budget | $0.12 | $0.50 | $0.21 |
| Qwen | Qwen3 8B | Budget | $0.12 | $0.45 | $0.20 |
| Qwen | Qwen3.6 35B A3B | Budget | $0.10 | $0.90 | $0.30 |
| Qwen | Qwen3.5-9B | Budget | $0.10 | $0.15 | $0.11 |
| Qwen | Qwen3 32B | Budget | $0.08 | $0.28 | $0.13 |
| Qwen | Qwen3.5-Flash | Budget | $0.07 | $0.26 | $0.11 |
| Qwen | Qwen3.7 Flash | Budget | $0.03 | $0.13 | $0.06 |
| SpaceXAI | Grok 4.6 | Mid | $2.00 | $6.00 | $3.00 |
| SpaceXAI | Grok 4.5 | Mid | $2.00 | $6.00 | $3.00 |
| SpaceXAI | Grok 4.3 | Mid | $1.25 | $2.50 | $1.56 |
| SpaceXAI | Grok 4.20 Multi-Agent | Mid | $1.25 | $2.50 | $1.56 |
| SpaceXAI | Grok 4.20 | Mid | $1.25 | $2.50 | $1.56 |
| SpaceXAI | Grok Build 0.1 | Mid | $1.00 | $2.00 | $1.25 |
*Blended = 3 input : 1 output, a common chat ratio. Prices are dynamic and fetched live. Cache misses are shown; cached input is typically ~10% of base rate on supported providers.
Pricing Data Source
Our calculator now fetches live pricing data directly from the OpenRouter API. This means the prices above are always up-to-date with the latest market rates for both input and output tokens.
What Is a Token, Anyway?
A token is the smallest unit of text a model reads or writes - roughly ¾ of a word. "The quick brown fox" ≈ 4 tokens. Pricing is quoted per 1 million tokens because real workloads (chat apps, agents, document processing) routinely consume millions of tokens a month.
Two things drive your bill:
- Input tokens - everything you send: your prompt, conversation history, retrieved context (RAG), system instructions.
- Output tokens - everything the model generates, including any hidden "thinking" tokens in reasoning models.
The rule of thumb: output is always more expensive than input - anywhere from 2× to 5×. A chatbot that generates long answers will burn far more budget on output than input, even though the prompt feels like the "big" part.
How to Cut Your Token Costs (Without Cutting Quality)
- Cache your context. Most providers price cached input at ~10% of the base rate. If the same system prompt + documents are sent repeatedly, caching can cut input cost by 90%.
- Batch non-urgent jobs. OpenAI and xAI offer ~50% off for async/batch processing. Perfect for summarization, classification, and backfill jobs.
- Right-size the model. Reserve flagships for reasoning-heavy work. Route routine tasks to faster, budget-friendly alternatives.
- Watch output tokens. Reasoning models burn hidden "thinking" tokens on output. Cap
max_tokensand tune prompts to answer concisely. - Go private for high volume. Past a few hundred million tokens a month, self-hosted open models (Qwen, DeepSeek, Llama, Mistral) on your own GPU typically undercut API pricing - Vistaran can run that for you.
FAQ
Why is output more expensive than input?
Output tokens are generated one at a time and can't be cached or parallelized the way input can. Reasoning models add hidden "thinking" tokens on the output side, further inflating the ratio.
What are cached input tokens?
When the same prompt prefix is sent repeatedly, the provider stores and reuses the computed representation, charging a fraction (~10%) of the normal input rate. This is the single biggest lever for RAG and chatbot cost.
Do these prices include a "thinking" / reasoning surcharge?
For most providers, thinking tokens are billed as output tokens at the standard output rate (Gemini and Grok bill them this way). Always check whether a reasoning model's quoted price includes or excludes thinking.
Is this calculator for API pricing or ChatGPT/Claude Pro subscriptions?
API pricing - the per-token rates you pay when building your own product on these models. Consumer subscriptions (ChatGPT Plus, Claude Pro) are flat monthly fees and aren't comparable per token.
LET'S CONNECT
Build on the Right Model - and the Right Stack
Picking a model is only step one. The bigger decision is whether you run it on a public API, on dedicated infrastructure, or on-premise for compliance. That's exactly what we engineer every day.
AI AGENT AS A SERVICEPRIVATE LLM HOSTINGAI AUTOMATION AUDIT
BUILT BY EXPERTS
Engineers shipping AI since 2017
Boomerr.ai, Scanview, and Compliance View in production
Secure, private deployments