OpenAI

o4-mini

o4-mini is a OpenAI o-series model from OpenAI, released in 2025-04. It costs $1.1 / $4.4 per 1M, has a 200k-token context window, and is best for cheap-reasoning, math, research. Last verified 2026-04-19.

Try o4-mini Docs Take the picker quiz

Spec sheet

Pricing

Input
$1.1 / 1M
Output
$4.4 / 1M
Cached input
$0.275 / 1M
Batch discount
50%
Free tier
No

Context & speed

Context window
200k tokens
Max output
100k tokens
Throughput
~60 tok/s
Time to first token
~1500 ms
Speed tier
slow

Capabilities

Tool use
Yes
Structured output
Yes
Prompt caching
Yes
Extended thinking
Yes
Vision input
Yes
Audio in / out
No
Fine-tuning
No

Deployment

Open weights
No
On-prem
No
HIPAA eligible
Yes
Zero retention
Yes
Regions
us, eu

Estimated monthly cost

Assumes typical token shape: 2k input, 600 output per call. Prompt caching is excluded from these figures.

10k calls/mo
$48.40
per month
100k calls/mo
$484.00
per month
1M calls/mo
$4.8k
per month

When to use o4-mini

Sweet spot

  • cheap reasoning
  • math
  • research
  • hardest coding

Known trade-offs

  • slow first-token
  • no streaming-first use cases

Works with

OpenAI SDKAzure OpenAIVercel AI SDKLangChainOpenRouter

Compare o4-mini to other models

vs GPT-5.4 Minivs Claude Haiku 4.5vs GLM-5.1vs Gemini 3 Flashvs Qwen 3.6 Plusvs Grok 4.20

FAQ — o4-mini

How much does o4-mini cost?

o4-mini costs $1.1 / $4.4 per 1M tokens on the OpenAI API. Cached input reads cost $0.275 per 1M, cutting the input bill by roughly 75% on repeat system prompts. The batch API offers a 50% discount for async workloads.

What is the context window of o4-mini?

o4-mini has a 200k-token context window with up to 100k tokens of output. That's enough for long reports, extended chat histories, or structured document analysis.

Does o4-mini have a free tier?

No — Paid-only.

Is o4-mini HIPAA / EU / on-prem friendly?

o4-mini is HIPAA-eligible, available in EU regions, and is API-only. Zero data retention is available for enterprise customers.

What is o4-mini best for?

o4-mini is best for cheap reasoning, math, research, hardest coding. Trade-offs to be aware of: slow first-token; no streaming-first use cases.

Which tools and SDKs work with o4-mini?

o4-mini integrates with OpenAI SDK, Azure OpenAI, Vercel AI SDK, LangChain, OpenRouter. Most major AI frameworks support it either natively or through OpenAI-compatible endpoints.