← Models

GPT-4o

OpenAI · GPT-4o

Widely deployed multimodal model kept as a legacy reference after retirement from ChatGPT defaults.

Part of GPT family · Other versions: GPT-5.5 Pro , GPT-5.5 , GPT-Rosalind , GPT-5.4 mini , GPT-5.4 nano , GPT-5.4-Pro , GPT-5.4 , GPT-5.3 , GPT-5.3-Codex , GPT-5.2-Codex , GPT-5.2-Pro , GPT-5.2 , GPT-5-Codex , GPT-5 mini , GPT-5 , GPT-5 nano , GPT-4.1 , GPT-4o mini Transcribe , GPT-4o mini TTS , GPT-4o Transcribe , GPT-4o mini
Type
multimodal
Context
128K tokens
Max Output
16K tokens
Status
legacy
Input
$2.5/1M tok
Output
$10/1M tok
API Access
Yes
License
proprietary
multimodal reasoning vision tool-use general-purpose
Released May 2024 · Updated March 27, 2026

Overview

Freshness note: Model capabilities, limits, and pricing can change quickly. This profile is a point-in-time snapshot last verified on March 27, 2026.

GPT-4o was OpenAI’s widely deployed multimodal model tier for mixed workloads across text, vision, and tool-enabled workflows. OpenAI retired GPT-4o from ChatGPT on February 13, 2026 while keeping API access available, and Help Center guidance says GPT-4o will be fully retired across Custom GPTs on April 3, 2026. That makes it a legacy compatibility route rather than a current default.

Capabilities

The model handles instruction-following, general analysis, structured outputs, and multimodal interpretation with strong consistency. It is still suitable for customer-facing copilots, workflow automation, and broad product integrations that were built around GPT-4o-era behavior.

Technical Details

OpenAI’s model card still lists GPT-4o with a 128K context window and 16,384 max output tokens. It remains one of the simpler older OpenAI multimodal compatibility options if you need a familiar model behavior and do not want to retune everything around GPT-5 generation changes.

Pricing & Access

OpenAI’s pricing docs still list GPT-4o at 2.50per1Minputtokens,2.50 per 1M input tokens, 1.25 cached input, and $10.00 per 1M output tokens. API availability remains active even though the ChatGPT-facing route has been retired.

Best Use Cases

Use GPT-4o when you still need compatibility with existing API integrations or established multimodal workflows. For new deployments, OpenAI’s current defaults have moved forward to GPT-5-family models and newer reasoning routes.

Comparisons

Compared with GPT-4o mini, GPT-4o usually offers higher quality on complex reasoning and multimodal tasks. Compared with GPT-5.2, GPT-4o is now the older route in OpenAI’s API lineup. Compared with Gemini 2.5 Flash, tradeoffs often depend on ecosystem fit and latency goals.