GPT-4
OpenAI's flagship model, GPT-4 is a large-scale multimodal language model capable of solving difficult problems with greater accuracy than previous models due to its broader general knowledge and advanced reasoning…
gpt-4Released 28 May 2023 (May 2023)Pricing
List prices at OpenAI, in US dollars per million tokens.
| Tier | Input / 1M | Output / 1M | |
|---|---|---|---|
| Standard | $30.00 | $60.00 |
What it costs a month
At OpenAI list prices, 30 days.
- Chat assistant5,000 requests a day, 2K in / 0.5K out$13,500
- RAG over documents2,000 requests a day, 8K in / 0.6K out$16,560
- Coding agent1,000 requests a day, 40K in / 2K out$39,600
- Overnight batch20,000 requests a day, 4K in / 0.4K out$86,400
Where to run it
1 host serve GPT-4. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
OpenAILab API gpt-4 | $30.00 / $60.00 | gpt-4 | |
GitHub CopilotSubscription gpt-4 | In the subscription | gpt-4 |
Show 3 more listings: regions, snapshots and variantsHide regional listings
| OpenAI | |||||||
| 0613 gpt-4-0613 | $30.00 / $60.00 | gpt-4-0613 | |||||
| 0613 · fine-tuning ft:gpt-4-0613 | $30.00 / $60.00 | ft:gpt-4-0613 | |||||
| GitHub Copilot | |||||||
| 0613 gpt-4-0613 | In the subscription | gpt-4-0613 | |||||
Capabilities
Core
- Streaming
- Function calling
- JSON mode
- Structured outputs
Multimodal
- Vision
- Image input
- Audio
- Image generation
- Video
Advanced
- Reasoning
- Tool calling
- MCP
- Prompt caching
- Batch API
- Embeddings
- Takes in
- Text
- Produces
- Text
- Features
- Prompt caching
Cheaper alternatives
Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.
Same model, cheaper host
No host sells it for less than OpenAI.
Smaller OpenAI models
- 57% less
- 87% less
- 89% less
Other frontier labs
- 57% less
- 89% less
- 92% less
- 79% less
Open-weight picks
- 89% less
- 92% less
- 95% less
- 95% less
Model line
Newer and older versions, at each lab's own price.
Call it
Using OpenAI's API.
from openai import OpenAI
client = OpenAI() # reads OPENAI_API_KEY
response = client.responses.create(model="gpt-4", input="Hello")
print(response.output_text)curl https://api.openai.com/v1/responses \
-H "Authorization: Bearer $OPENAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "gpt-4", "input": "Hello"}'from litellm import completion
response = completion(
model="openai/gpt-4",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
How much does GPT-4 cost?
On OpenAI, GPT-4 costs $30.00 per million input tokens and $60.00 per million output tokens.
Which APIs offer GPT-4?
GPT-4 runs on 1 API: OpenAI.
What is GPT-4's context window?
8K tokens, with up to 4K tokens in one reply.
Is GPT-4 being retired?
No retirement date is scheduled for it on any API we track.
Is there a newer version of GPT-4?
Yes. GPT-5.6 is the newest model in this line.
What are cheaper alternatives to GPT-4?
Models that cost less and keep most of what it can do: GPT-6.2 Astra on OpenAI (57% less), Claude Fable 5.2 on Anthropic (57% less), Gemini 4 Carbon on Google AI Studio (89% less) and Kimi K3.1 Pro on Databricks (89% less). Savings are on a mix of 3 input tokens to 1 output token.
When was GPT-4 released?
OpenAI released it on 28 May 2023.
What is GPT-4's API model id?
Use gpt-4 with OpenAI.
Sources
Raw catalog entry for gpt-4 on OpenAI
{
"deprecation_date": "2026-10-23",
"deprecation_source": "https://developers.openai.com/api/docs/deprecations",
"successor_model": "gpt-5.6-sol",
"input_cost_per_token": 0.00003,
"litellm_provider": "openai",
"max_input_tokens": 8192,
"max_output_tokens": 4096,
"mode": "chat",
"output_cost_per_token": 0.00006,
"fedramp_uplift_multiplier": 1.1,
"supports_function_calling": true,
"supports_prompt_caching": true,
"supports_system_messages": true,
"supports_tool_choice": true
}