LANGUAGE MODEL OpenAI Last updated:

GPT-4o (Aug '24)

GPT-4o (Aug '24) is an API model from OpenAI. It’s positioned for general text tasks—work that benefits from iteration, not just one-shot answers.

Intelligence
Below avg
Speed
Medium
108 tok/s output
Cost
High
$5.00 in / $15.00 out
Context
128K
Up to 128,000 tokens
How are Intelligence, Speed & Cost bucketed?
Intelligence and Speed buckets are percentile ranks on Artificial Analysis. Cost buckets are fixed dollar thresholds keyed off output-token price ($/M out).
Intelligence
  • Top 1%≤ 1%
  • Top 5%≤ 5%
  • Top 10%≤ 10%
  • Good≤ 25%
  • Medium≤ 50%
  • Below avg> 50%
Speed
  • Top 1%≥ 345 tok/s
  • Top 5%≥ 237 tok/s
  • Top 10%≥ 196 tok/s
  • Good≥ 146 tok/s
  • Medium≥ 90 tok/s
  • Slow< 90 tok/s
Cost
  • Freeopen weights · self-host
  • Low< $1 / M out
  • Moderate$1–5 / M out
  • High≥ $5 / M out

Core Capabilities

Long Documents
Handles entire codebases, books, and multi-doc RAG.

Context Window

128k tokens
≈ 98 pages
4k Chat 聊天
32k Long docs 长文档
128k This model 本模型
400k Multi-doc 多文档
1M Codebase 整个代码库
10M

Availability

API
Available
Product / App
Not available
Open Source
Not released
Enterprise
Contact sales

Pricing Model

Pay per token
Input and output billed separately.
Pay-per-token

Capability / Performance

Where this model sits relative to the middle 60% of models in the tree. All scores are 0–10 (higher is better).

Lower 20% Upper 80% This model
Reasoning
AA Intelligence Index · scaled to 10
1.7
5.6
2.7
Coding
SciCode · scaled to 10
1.8
4.3
3.7
Agentic tasks
Terminal-Bench Hard · scaled to 10
0.2
3.6
0.8
Context / memory
Context window size · log-scaled
6.0
9.0
6.0
Cost efficiency
Input price ($/M tokens) · cheaper scores higher
6.2
10.0
3.8
Lower 20% 20th percentile — 20% of models score below this This model Where the current model lands Upper 80% 80th percentile — only 20% of models score above this Percentile boundaries are computed across every model in the tree that reports the underlying benchmark for each capability.

What it feels like

  • First end-to-end omni model — text, vision, audio share one neural net (not stitched pipelines)
  • 2x faster, half the price, and 5x higher rate limits than GPT-4 Turbo
  • Native voice-to-voice latency around 320ms — close to human conversational rhythm (210ms)

Best use cases

  • Voice-first applications and real-time multimodal interfaces
  • Cost-sensitive bulk inference where GPT-4-class quality is enough
  • Image understanding workflows — strong vision pipeline at API price

Tools to try

Not ideal for

  • Hard reasoning, math, or research — o1/o3/GPT-5 are the right tier
  • Frontier-leaderboard coding (Claude 3.5 Sonnet+ outscored GPT-4o on SWE-bench by mid-2024)

Model Evolution

GPT is OpenAI's language model family.

View full evolution tree →