This model was retired on 2027-09-21. The pricing below is the last-known rate, kept for migration reference.
GPT-5.4 MiniDeprecated
VisionTool useReasoningPrompt cachingStructured outputWeb search
Context
272K
Max output
128K
Serving providers
4
Cheapest input
$0.75 /1M
Prices are on-demand list rates. Enterprise commitments (committed-use discounts, provisioned throughput, negotiated rates) can be materially lower.
Benchmarks
independent evaluations · source: ARC Prize and Artificial AnalysisSpeed & latency
Output speed
164 t/s
tokens / second
Time to first token
10.48s
latency
Headline indices
Intelligence Index
40.9
of 100
Coding Index
56.1
of 100
Agentic Index
31.5
of 100
Individual benchmarks · click a row for effort variants
| Benchmark | Best score | Effort | Cost / task | |||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Aa Analyst Agent | 0.1 | default | — | |||||||||||||||||||||
| AA-Briefcase | 10% | — | — | |||||||||||||||||||||
| APEX Agents | 28% | default | — | |||||||||||||||||||||
| ARC-AGI-2 | 19% | xhigh | $0.750 | |||||||||||||||||||||
| ||||||||||||||||||||||||
| CritPt | 10% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| GDPval | 1171.8 | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| GPQA Diamond | 87% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Harvey Lab | 0.6 | default | — | |||||||||||||||||||||
| Humanity's Last Exam | 28% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| IFBench | 73% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| IT-Bench SRE | 35% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Long-Context Reasoning | 73% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Mlcr Overall | 0.0 | default | — | |||||||||||||||||||||
| MMMU-Pro | 73% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Omniscience | -18.9 | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Omniscience Accuracy | 0.4 | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Omniscience Non Hallucination | 0.1 | medium | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| SciCode | 50% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| TerminalBench Hard | 52% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
| TerminalBench v2.1 | 59% | default | — | |||||||||||||||||||||
| τ-bench Banking | 26% | default | — | |||||||||||||||||||||
| τ²-bench | 83% | default | — | |||||||||||||||||||||
| ||||||||||||||||||||||||
Pricing by serving provider
last-known · per 1M tokens · click a provider to expand| Serving provider | Input /1M | Output /1M | Endpoints | |||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Databricks | $0.75 | $4.50 | 1 | |||||||||||||||||||||
| ||||||||||||||||||||||||
| Azure | $0.75 | $4.50 | 1 | |||||||||||||||||||||
| ||||||||||||||||||||||||
| OpenAI | $0.75 | $4.50 | 1 | |||||||||||||||||||||
| ||||||||||||||||||||||||
| OpenRouter | $0.75 | $4.50 | 1 | |||||||||||||||||||||
| ||||||||||||||||||||||||
Price history
input + output $/1M since we started tracking Input Output
Cost calculator
estimate your monthly spend on this model$510
estimated / month
Model IDs
copy the exact identifier for your platformdatabricks/databricks-gpt-5-4-miniazure/gpt-5.4-minigpt-5.4-miniopenai/gpt-5.4-mini
Frequently asked questions
GPT-5.4 Mini pricing, context and availabilityIs GPT-5.4 Mini still available?
GPT-5.4 Mini was retired on 2027-09-21. The pricing on this page is the last-known rate, kept for migration reference.
How much did GPT-5.4 Mini cost?
GPT-5.4 Mini's last-known pricing, before it was retired on 2027-09-21, was $0.75 per 1M input tokens and $4.50 per 1M output tokens.
What was the context window of GPT-5.4 Mini?
GPT-5.4 Mini had a 272K token context window and could return up to 128K output tokens.
What could GPT-5.4 Mini do?
GPT-5.4 Mini supported image input (vision), tool use, reasoning, prompt caching, structured output and web search.
Other OpenAI models
compare pricing across the OpenAI lineup Pricing verified against LiteLLM + OpenRouter + provider pages · refreshed daily
Every weekday
AI moves fast. Here's your debrief.
News, analysis, tools, and more.
For people who build with AI