MDL
GPT-5.4
OpenAI
Verified against Artificial Analysis · Jul 13, 2026
Capability
95
percentile index
Price /M
$2.5 / $15
in / out
Context
1.1M
tokens
Avg score
69.0
7 benchmarks
AA Intelligence
51.4
index
Speed
161 tok/s
TTFT 1m 44s
Released
Mar 5, 2026
About
GPT-5.4 is a proprietary model in the GPT family from OpenAI. Benchmarked on 7 evals, averaging 69.0.
Benchmark scores · 7
avg 69.0 — higher is betterThe research behind this model
all papers mentioning it →Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models▲ 165 on HF · Jun 9, 2026AgentDoG 1.5: A Lightweight and Scalable Alignment Framework for AI Agent Safety and Security▲ 142 on HF · May 28, 2026PlanBench-XL: Evaluating Long-Horizon Planning of LLM Tool-Use Agents in Large-Scale Tool Ecosystems▲ 95 on HF · Jun 21, 2026S-Agent: Spatial Tool-Use Elicits Reasoning for Spatial Intelligence▲ 39 on HF · Jun 18, 2026SkillOpt-Lite: Better and Faster Agent Self-evolution via One Line of Vibe▲ 29 on HF · Jul 3, 2026
Mentions matched by name in title/abstract — from the arXiv + HF daily corpus.
Performance & price
Output speed161 tok/s
Latency (TTFT)1m 44s
Price in / out$2.5 / $15 per 1M
Blended price$5.63 / 1M
Coding index71.1
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$15/M