MDL
GPT-5.2
OpenAI
Verified against Artificial Analysis · Jul 13, 2026
Capability
95
percentile index
Price /M
$1.75 / $14
in / out
Context
400K
tokens
Avg score
73.3
10 benchmarks
AA Intelligence
42.2
index
Speed
81 tok/s
TTFT 1m 46s
Released
Dec 11, 2025
About
GPT-5.2 is a proprietary model in the GPT family from OpenAI. Benchmarked on 10 evals, averaging 73.3.
Benchmark scores · 10
avg 73.3 — higher is betterThe research behind this model
all papers mentioning it →SkillAdaptor: Self-Adapting Skills for LLM Agents from Trajectories▲ 35 on HF · May 31, 2026Exploring Autonomous Agentic Data Engineering for Model Specialization▲ 23 on HF · May 28, 2026CausaLab: A Scalable Environment for Interactive Causal Discovery Toward AI Scientists▲ 18 on HF · May 25, 2026DEI: Diversity in Evolutionary Inference for Quality-Diversity Search▲ 14 on HF · May 26, 2026RepoRescue: An Empirical Study of LLM Agents on Whole-Repository Compatibility Rescue▲ 4 on HF · Jul 1, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed81 tok/s
Latency (TTFT)1m 46s
Price in / out$1.75 / $14 per 1M
Blended price$4.81 / 1M
Math index99.0
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$14/M