OpenAI · Model
GPT-4o pricing, context window and specs
GPT-4o costs $2.50 per million input tokens and $10 per million output tokens, and reads up to 128,000 tokens per request. By blended price it is number 6 of OpenAI’s 10 current models here, counting from the cheapest.
- Input / 1M
- $2.50
- Output / 1M
- $10
- Cached / 1M
- $1.25
- Context
- 128,000
- Max output
- 16,384
- Released
- 2024-05-13
Estimate your cost with GPT-4oCompare all OpenAI modelsCount tokens for GPT-4o
Pricing
GPT-4o API pricing
| Price per 1M tokens | Input | Output |
|---|---|---|
| Standard | $2.50 | $10 |
| Cached input (read) | $1.25 | – |
| Batch API | $1.25 | $5 |
Output costs 4.0× the input price, so long replies drive the bill more than long prompts. If your requests share a long fixed prefix, such as a system prompt or tool definitions, caching cuts that part to 50% of the normal input price.
Examples
What GPT-4o costs in practice
| Workload | Tokens in / out | Requests/day | Per request | Per month |
|---|---|---|---|---|
| Support chatbot50% cached | 1,500 / 400 | 1,000 | $0.00681 | $207.21 |
| RAG app20% cached | 6,000 / 500 | 500 | $0.0185 | $281.35 |
| Coding agent80% cached | 40,000 / 2,000 | 200 | $0.08 | $486.67 |
| Document summariserbatch | 8,000 / 600 | 300 | $0.013 | $118.63 |
Change any number in the cost calculator.
Context
Context window
GPT-4o accepts up to 128,000 tokens per request, roughly 96,000 English words or 192 pages. That budget is shared between your prompt and the reply, and a single reply is capped at 16,384 tokens.
Filling the whole window with one prompt costs about $0.32 at list price, so for repeated long-document work, retrieval (sending only the relevant parts) or prompt caching is usually much cheaper.
Features
Capabilities
- Image input
- Yes
- Tool / function calling
- Yes
- Reasoning mode
- No
- Prompt caching
- Yes
- Structured output
- Yes
- Open weights
- No
- Accepts
- text, image, file
- Knowledge cutoff
- 2023-10-31
As declared by the provider’s API. It shows a feature exists, not how well it works.
API
GPT-4o model ID
- OpenRouter
openai/gpt-4o- OpenAI API
gpt-4o(alias)gpt-4o-2024-11-20(pinned)gpt-4o-2024-08-06(pinned)
OpenAI lists GPT-4o as current. Every GPT model ID, with aliases and snapshots.
OpenAI
Where GPT-4o sits in OpenAI’s line-up
Ranked by blended price (three parts input to one part output, $4.375 per 1M for GPT-4o), it is number 6 of OpenAI’s 10 current models here, counting from the cheapest. One step down, GPT-6.1 Sol costs $4 blended (9% less). Unlike GPT-4o, it reads up to 1,050,000 tokens (128,000 here), caps a reply at 128,000 tokens (16,384 here) and has a reasoning mode. One step up, GPT-5.6 Terra costs $4.50 (1× as much). Unlike GPT-4o, it reads up to 1,050,000 tokens (128,000 here), caps a reply at 128,000 tokens (16,384 here) and has a reasoning mode. See every OpenAI model and price
| Model | Input / 1M | Output / 1M | Context | Released |
|---|---|---|---|---|
| gpt-oss-120b | $0.037 | $0.17 | 131,072 | 2025-08-05 |
| GPT-6 Luna | $0.10 | $0.50 | 1,050,000 | 2026-09-22 |
| GPT-4o-mini | $0.15 | $0.60 | 128,000 | 2024-07-18 |
| GPT-5.4 Mini | $0.75 | $4.50 | 400,000 | 2026-03-17 |
| GPT-6.1 Sol | $2 | $10 | 1,050,000 | 2026-09-29 |
| GPT-4o (this page) | $2.50 | $10 | 128,000 | 2024-05-13 |
| GPT-5.6 Terra | $2 | $12 | 1,050,000 | 2026-07-09 |
| GPT-5.5 | $5 | $30 | 1,050,000 | 2026-04-24 |
| GPT-6 Astra | $10 | $50 | 1,050,000 | 2026-09-04 |
| GPT-5.5 Pro | $30 | $180 | 1,050,000 | 2026-04-24 |
OpenAI’s current models with a page here, cheapest first. Older and superseded models are listed on the provider page and in the comparison table.
Save
Cheaper alternatives
- GLM 4.6VZ.ai90% cheaper
- Muse Glimmer 30BMeta88% cheaper
- Qwen3.7 PlusQwen87% cheaper
- Command A+Cohere86% cheaper
- Seed-2.0-LiteByteDance Seed84% cheaper
Current models from major providers, one per provider, that keep image input, tool calling, a usable context window and output limit. Compared by blended price.
Compare
Similarly priced models
- GLM 5.3 PrimeZ.ai$2.80 / $8.80
- Gemini 3.1 Pro PreviewGoogle$2 / $12
- Claude Sonnet 5.5Anthropic$2 / $10
- Nova Premier 1.0Amazon$2.50 / $12.50
- Qwen3.8 Max PrimeQwen$4 / $12
Current models from other major providers, closest blended price.
FAQ
GPT-4o questions
How much does GPT-4o cost?
GPT-4o costs $2.50 per million input tokens and $10 per million output tokens, with cached input at $1.25. A typical chatbot reply (1,500 tokens in, 400 out) costs about $0.00775. Use the LLM cost calculator for your own workload.
What is GPT-4o’s context window?
128,000 tokens, about 96,000 English words or 192 printed pages, shared between your prompt and the reply. A single reply can be up to 16,384 tokens.
Does GPT-4o support prompt caching and batch requests?
Yes, cached input is billed at $1.25 per million tokens (50% of the normal input price). A batch API is available at $1.25 input and $5 output per million tokens.
What are cheaper alternatives to GPT-4o?
With the same essentials (image input and tool calling), the cheapest options are GLM 4.6V (90% cheaper), Muse Glimmer 30B (88% cheaper), Qwen3.7 Plus (87% cheaper). Cheaper doesn’t mean equivalent, so test them on your own prompts.
What is the API model ID for GPT-4o?
On OpenRouter it is openai/gpt-4o. In OpenAI’s own API, use gpt-4o; OpenAI lists the model as current. The GPT model ID list has every alias and snapshot. Pass the ID as the model value in each request.
Is GPT-4o open source?
No, GPT-4o is only available through OpenAI’s API and partners; its weights aren’t published.