MDL
GPT-5.4 nano
OpenAI
Verified against Artificial Analysis · Jul 13, 2026
Capability
87
percentile index
Price /M
$0.2 / $1.25
in / out
Context
400K
tokens
Avg score
59.3
7 benchmarks
AA Intelligence
38.2
index
Speed
176 tok/s
TTFT 2.6s
Released
Mar 17, 2026
About
GPT-5.4 nano is a proprietary model in the GPT family from OpenAI. Benchmarked on 7 evals, averaging 59.3.
Benchmark scores · 7
avg 59.3 — higher is betterThe research behind this model
all papers mentioning it →SkillOpt-Lite: Better and Faster Agent Self-evolution via One Line of Vibe▲ 29 on HF · Jul 3, 2026Towards Retrieving Interaction Spaces for Agentic Search▲ 4 on HF · Jun 5, 2026Training Deliberative Monitors for Black-Box Scheming DetectionMay 28, 2026A Two-Phase Stability Study of LLM Judges and Bar Council Examiners on Thai Bar-Exam Free-Form EssaysMay 25, 2026Agentic Systems as Boosting Weak Reasoning ModelsMay 13, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed176 tok/s
Latency (TTFT)2.6s
Price in / out$0.2 / $1.25 per 1M
Blended price$0.463 / 1M
Coding index56.1
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$1.25/M