Checked daily
AI API pricing for every current model
What 59 models from OpenAI, Anthropic, Google and xAI cost per million tokens, read every day from each provider's own pricing page. Prices are in US dollars per 1 million tokens. "A chat reply" is one support chat answer: 1,500 tokens in, 300 out.
Cheapest for a chat reply
- GPT-5 nano (OpenAI): $0.00019 per reply, $1.95 per 10,000
- GPT-4.1 nano (OpenAI): $0.00027 per reply, $2.70 per 10,000
- Gemini 2.5 Flash-Lite (Google): $0.00027 per reply, $2.70 per 10,000
- GPT-6 Luna (OpenAI): $0.00030 per reply, $3.00 per 10,000
- Claude Haiku 5.5 (Anthropic): $0.00030 per reply, $3.00 per 10,000
Cheapest isn't the same as good enough: that depends on what each request asks for.
Announced price changes
Gemini 3.8 Flash. Input: $0.75 per 1M tokens through December 31, 2026, then $1.50 from January 1, 2027.
Gemini 3.8 Flash. Output: $3.75 per 1M tokens through December 31, 2026, then $7.50 from January 1, 2027.
Gemini 3.6 Flash. Input: $0.75 per 1M tokens through December 31, 2026, then $1.50 from January 1, 2027.
Gemini 3.6 Flash. Output: $3.75 per 1M tokens through December 31, 2026, then $7.50 from January 1, 2027.
OpenAI
Newest first. Source: OpenAI's pricing page.
| Model | Input | Output | Cached input | A chat reply | Context |
|---|---|---|---|---|---|
| GPT-6.1 Sol | $2 | $10 | $0.10 | $0.0060 | 1.05M |
| GPT-6 Sol | $2 | $10 | $0.20 | $0.0060 | 1.05M |
| GPT-6 Luna | $0.10 | $0.50 | $0.01 | $0.00030 | 1.05M |
| GPT-6 Astra | $10 | $50 | $1 | $0.03 | 1.05M |
| GPT-5.6 Sol | $4 | $20 | $0.40 | $0.012 | 1.05M |
| GPT-5.6 Terra | $2 | $12 | $0.20 | $0.0066 | 1.05M |
| GPT-5.6 Luna | $0.20 | $1.20 | $0.02 | $0.00066 | 1.05M |
| GPT-5.5 Pro | $30 | $180 | - | $0.099 | 1.05M |
| GPT-5.5 | $5 | $30 | $0.50 | $0.017 | 1.05M |
| GPT-5.4 mini | $0.75 | $4.50 | $0.075 | $0.0025 | 400K |
| GPT-5.4 nano | $0.20 | $1.25 | $0.02 | $0.00068 | 400K |
| GPT-5.4 Pro | $30 | $180 | - | $0.099 | 1.05M |
| GPT-5.4 | $2.50 | $15 | $0.25 | $0.0083 | 1.05M |
| GPT-5.2 Pro | $21 | $168 | - | $0.082 | 400K |
| GPT-5.2 | $1.75 | $14 | $0.175 | $0.0068 | 400K |
| GPT-5.1 | $1.25 | $10 | $0.125 | $0.0049 | 400K |
| GPT-5 Pro | $15 | $120 | - | $0.059 | 400K |
| GPT-5 | $1.25 | $10 | $0.125 | $0.0049 | 400K |
| GPT-5 mini | $0.25 | $2 | $0.025 | $0.00097 | 400K |
| GPT-5 nano | $0.05 | $0.40 | $0.005 | $0.00019 | 400K |
| o3-pro | $20 | $80 | - | $0.054 | 200K |
| o3 | $2 | $8 | $0.50 | $0.0054 | 200K |
| o4-mini | $1.10 | $4.40 | $0.275 | $0.0030 | 200K |
| GPT-4.1 | $2 | $8 | $0.50 | $0.0054 | 1.05M |
| GPT-4.1 mini | $0.40 | $1.60 | $0.10 | $0.0011 | 1.05M |
| GPT-4.1 nano | $0.10 | $0.40 | $0.025 | $0.00027 | 1.05M |
| o1-pro | $150 | $600 | - | $0.405 | 200K |
| o3-mini | $1.10 | $4.40 | $0.55 | $0.0030 | 200K |
| o1 | $15 | $60 | $7.50 | $0.041 | 200K |
| GPT-4o mini | $0.15 | $0.60 | $0.075 | $0.00040 | 128K |
| GPT-4o | $2.50 | $10 | $1.25 | $0.0067 | 128K |
Anthropic
Newest first. Source: Anthropic's pricing page.
| Model | Input | Output | Cached input | A chat reply | Context |
|---|---|---|---|---|---|
| Claude Haiku 5.5 | $0.10 | $0.50 | $0.01 | $0.00030 | 1M |
| Claude Sonnet 5.5 | $2 | $10 | $0.10 | $0.0060 | 1M |
| Claude Opus 5.5 | $4 | $20 | $0.20 | $0.012 | 1M |
| Claude Fable 5.1 | $10 | $50 | $0.25 | $0.03 | 1M |
| Claude Opus 5 | $5 | $25 | $0.50 | $0.015 | 1M |
| Claude Sonnet 5 | $2 | $10 | $0.20 | $0.0060 | 1M |
| Claude Fable 5 | $10 | $50 | $1 | $0.03 | 1M |
| Claude Opus 4.8 | $5 | $25 | $0.50 | $0.015 | 1M |
| Claude Opus 4.7 | $5 | $25 | $0.50 | $0.015 | 1M |
| Claude Sonnet 4.6 | $3 | $15 | $0.30 | $0.0090 | 1M |
| Claude Opus 4.6 | $5 | $25 | $0.50 | $0.015 | 1M |
| Claude Opus 4.5 | $5 | $25 | $0.50 | $0.015 | 200K |
| Claude Haiku 4.5 | $1 | $5 | $0.10 | $0.0030 | 200K |
| Claude Sonnet 4.5 | $3 | $15 | $0.30 | $0.0090 | 1M |
Newest first. Source: Google's pricing page.
| Model | Input | Output | Cached input | A chat reply | Context |
|---|---|---|---|---|---|
| Gemini 3.8 Flash | $0.75 | $3.75 | $0.075 | $0.0022 | 1.05M |
| Gemini 3.6 Flash | $0.75 | $3.75 | $0.075 | $0.0022 | 1.05M |
| Gemini 3.5 Flash-Lite | $0.30 | $2.50 | $0.03 | $0.0012 | 1.05M |
| Gemini 3.1 Flash-Lite | $0.25 | $1.50 | $0.025 | $0.00082 | 1.05M |
| Gemini 3.1 Pro Preview | $2 | $12 | $0.20 | $0.0066 | 1.05M |
| Gemini 3 Flash Preview | $0.50 | $3 | $0.05 | $0.0016 | 1.05M |
| Gemini 2.5 Flash-Lite | $0.10 | $0.40 | $0.01 | $0.00027 | 1.05M |
| Gemini 2.5 Pro | $1.25 | $10 | $0.125 | $0.0049 | 1.05M |
| Gemini 2.5 Flash | $0.30 | $2.50 | $0.03 | $0.0012 | 1.05M |
xAI
Newest first. Source: xAI's pricing page.
| Model | Input | Output | Cached input | A chat reply | Context |
|---|---|---|---|---|---|
| Grok 4.7 | $2 | $6 | $0.50 | $0.0048 | 500K |
| Grok 4.6 | $2 | $6 | $0.50 | $0.0048 | 500K |
| Grok 4.5 | $2 | $6 | $0.30 | $0.0048 | 500K |
| Grok Build 0.1 | $1 | $2 | $0.20 | $0.0021 | 256K |
| Grok 4.3 | $1.25 | $2.50 | $0.20 | $0.0026 | 1M |
Head-to-head comparisons
- Claude Sonnet 5.5 vs GPT-6 Sol
- Claude Haiku 5.5 vs GPT-6 Luna
- Claude Haiku 5.5 vs Gemini 3.8 Flash
- GPT-6 Luna vs Gemini 3.8 Flash
- Claude Opus 5.5 vs GPT-6 Astra
- Claude Sonnet 5.5 vs Claude Opus 5.5
- Claude Haiku 5.5 vs Claude Haiku 4.5
- Claude Sonnet 5.5 vs Claude Sonnet 5
- GPT-6 Sol vs GPT-5.6 Sol
- Gemini 3.8 Flash vs Gemini 2.5 Flash
- Grok 4.7 vs Claude Sonnet 5.5
- GPT-5.4 mini vs Claude Haiku 5.5
- GPT-4o mini vs GPT-6 Luna
- GPT-4o vs GPT-6 Sol
- Gemini 3.8 Flash vs GPT-5.4 mini
Paying flagship prices for every request?
Most apps send simple requests to the same strong model as the hard ones. PromptUnit checks each request your app sends and uses a cheaper model only when it is just as good for that request. You pay 20% of what it saves, and nothing if it saves nothing.
Checked October 14, 2026. Prices are read every day from OpenAI's pricing page, Anthropic's pricing page, Google's pricing page and xAI's pricing page. Context window, inputs and release date come from OpenRouter's public model list.
These are the standard list prices on each provider's own API. Batch, flex and priority tiers, cloud platforms such as Amazon Bedrock or Google Vertex AI, and regional endpoints are priced differently. Request costs above count input and output tokens only, not cache, search or image fees.