GPT-4o (2024-08-06)

openai/gpt-4o-2024-08-06
VisionToolsJSON
by OpenAI · 2024-08-06

The 2024-08-06 version of GPT-4o offers improved performance in structured outputs, with the ability to supply a JSON schema in the respone_format. Read more [here](https://openai.com/index/introducing-structured-outputs-in-the-api/). GPT-4o ("o" for "omni") is...

ctx128K tokens
Max output16.4K
Inputtext + image + file
Outputtext
p50 TTFT2.00 s
INPUT$2.50/ 1M tokens
OUTPUT$10.00/ 1M tokens
p50 TTFT2.00 s7d
p95 TTFT7.61 s7d
TRAFFIC207.0Ktokens / 7d

GPT-4o (2024-08-06) is a version of OpenAI's GPT-4o model, released in August 2024. It is a multimodal large language model that processes text, image, and file inputs. The model is accessed through…

What is GPT-4o (2024-08-06)?

Who is this model for?

What input modalities does it support?

Key specifications at a glance

Code samples

Call from any SDK

OpenAI-compatible — keep the SDK you already use

  • OpenAI SDKhttps://api.orcarouter.ai/v1
from openai import OpenAI

client = OpenAI(
    base_url="https://api.orcarouter.ai/v1",
    api_key="$ORCAROUTER_API_KEY",
)

response = client.chat.completions.create(
    model="openai/gpt-4o-2024-08-06",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Supported parameters

  • frequency_penalty
  • logit_bias
  • logprobs
  • max_completion_tokens
  • max_tokens
  • n
  • parallel_tool_calls
  • prediction
  • presence_penalty
  • response_format
  • seed
  • service_tier
  • stop
  • stream
  • structured_outputs
  • temperature
  • tool_choice
  • tools
  • top_logprobs
  • top_p
  • web_search_options

Pricing

Input / 1M tokens$2.50
Output / 1M tokens$10.00
Cache read / 1M$1.25
CurrencyUSD

Cost calculator

Tokens / month10MM
Input share70%%
Estimated / month $47.50 · With prompt caching $43.13

Estimate based on list price

Token & cost estimator

Input tokens: 20Cost per request: $0.005050

Estimate only — actual token counts depend on the provider's tokenizer.

Performance

p50 TTFT
2.00 s
Output speed
477 tok/s
p95 TTFT
7.61 s
Error rate
0%

Public benchmarks

16.6
AA Coding
Better than 12% of models compared
#108 of 123
9.6
AA Intelligence
Better than 10% of models compared
#110 of 125
AIME
11.7
GPQA Diamond
52.1
Humanity's Last Exam
2.9
IFBench
36.0
LiveCodeBench
31.7
Long-Context Recall
35.0
MATH-500
79.5
SciCode
33.1
TerminalBench Hard
8.3
τ²-Bench
28.9
Source: artificialanalysis.ai

How it compares

GPT-4o (2024-08-06)GPT-5.2 ProGPT-5.4 ProGPT-5.5
Input $/M$2.50$21.00$60.00$5.00
Output $/M$10.00$168.00$270.00$30.00
Context128K400K1.1M
Quality3/1010/1010/1010/10
Compare side-by-sideCompare side-by-sideCompare side-by-sideCompare side-by-side

FAQ

What is the cost per token for GPT-4o (2024-08-06) on OrcaRouter?
Pricing is $2.50 per 1 million input tokens and $10.00 per 1 million output tokens, billed at OpenAI's provider rate with zero markup from OrcaRouter. Image and file inputs incur additional token costs per OpenAI's policy.
What is the context window and max output tokens?
The context window is 128,000 tokens. The maximum output is 16,384 tokens. This allows processing of long documents or conversations and generating substantial responses.
What are the main strengths of this model?
It offers strong mathematical reasoning (79.5 on MATH-500), multimodal input (text, images, files), a large context window, and OpenAI's high-quality generation. It is well-suited for complex reasoning and vision tasks.
How does this model compare to a cheaper model like GPT-4o mini?
GPT-4o (2024-08-06) is more expensive but generally more capable in reasoning and multimodal tasks. GPT-4o mini is cheaper ($0.15/$0.60 per 1M tokens) and sufficient for simple tasks. Choose based on task complexity and budget.
What data handling policies apply?
Your data is sent to OpenAI for processing. OrcaRouter does not store prompts. OpenAI's data usage policies apply. Use OrcaRouter's multi-provider routing to direct requests to providers that meet your compliance requirements.
Can I use this model with the OpenAI Python client?
Yes, by setting the base URL to https://api.orcarouter.ai/v1 and using model ID "openai/gpt-4o-2024-08-06". Your existing code for ChatCompletion, streaming, and function calling works without changes.
How do I send image inputs via the API?
Include the image as a base64-encoded string or URL in the content array of a message, using the format "image_url" with detail parameter. The model interprets images for analysis. This follows OpenAI's multimodal API format.
What is the maximum output that can be generated per request?
The maximum output is 16,384 tokens. Use the max_tokens parameter to control the response length. Be aware that longer outputs increase cost and latency.
Is this model available for streaming?
Yes, when using the OpenAI-compatible API, you can set stream=True to receive responses incrementally. This works with OrcaRouter's endpoint.
Can I use this model for free or with a trial?
The model is billed at provider rates. OrcaRouter may offer a free tier or trial credits—check their website for current offers. Usage beyond any free credits incurs the standard per-token costs.

Embed this badge

OpenAI: GPT-4o (2024-08-06)$2.50/M in2000ms p50via OrcaRouter
HTML <a href="https://www.orcarouter.ai/models/openai/gpt-4o-2024-08-06" target="_blank"> <img src="https://www.orcarouter.ai/embed/openai/gpt-4o-2024-08-06.svg" alt="OpenAI: GPT-4o (2024-08-06) on OrcaRouter" /> </a>
Markdown [![OpenAI: GPT-4o (2024-08-06)](https://www.orcarouter.ai/embed/openai/gpt-4o-2024-08-06.svg)](https://www.orcarouter.ai/models/openai/gpt-4o-2024-08-06)

Model card as data

GET /api/public/models/openai/gpt-4o-2024-08-06Open
Machine-readable:/llms.txt/llms-full.txt