MDL
GPT-4o
OpenAI
Verified against Artificial Analysis · Jul 13, 2026
Capability
50
percentile index
Price /M
$2.5 / $10
in / out
Context
128K
tokens
Avg score
46.4
12 benchmarks
AA Intelligence
9.6
index
Speed
115 tok/s
TTFT 0.6s
Released
Aug 6, 2024
About
GPT-4o is a proprietary model in the GPT-4o family from OpenAI. Benchmarked on 12 evals, averaging 46.4.
proprietarytextvision
Benchmark scores · 12
avg 46.4 — higher is betterThe research behind this model
all papers mentioning it →SDR: Set-Distance Rewards for Radiology Report Generation▲ 3 on HF · May 30, 2026Send a SCOUT First: Pre-hoc Reasoning for Adaptive Detector Allocation in Prompt-Injection Defense▲ 1 on HF · May 29, 2026The Chain Holds, the Answer Folds: Trace-Answer Dissociation in Reasoning Models Under Adversarial Pressure▲ 1 on HF · May 27, 2026Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone151 cites · Apr 22, 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution76 cites · Sep 18, 2024
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed115 tok/s
Latency (TTFT)0.6s
Price in / out$2.5 / $10 per 1M
Blended price$4.38 / 1M
Median across API providers, via Artificial Analysis. Blended = 3:1 input:output. For reasoning models, latency includes thinking time. Methodology →
History
full ledger →Capability over time
Append-only ledger — every observed change to this model's numbers.
API providers
Artificial Analysis$10/M
OpenAI128K ctx · $10/M