OpenAI o4-mini
BalancedLast verified: June 2026· OpenAI OpenAI o4-mini pricing
Smaller reasoning model. Reasoning tokens billed as output — confirm current pricing with OpenAI.
- Input / 1M
- $1.10
- Output / 1M
- $4.40
- Context
- 200K tokens
- Provider
- OpenAI
Pricing verified June 2026. Prices change frequently. Always confirm against the provider’s official pricing page before relying on these figures for budgeting. Official pricing →
When to choose OpenAI o4-mini
Reasoning-model behaviour at a mid-table price: $1.10 in and $4.40 out, with a 200K window and the billing rule every reasoning row carries — reasoning tokens land on the output meter. That makes it a sensible default for reasoning-heavy routes running at volume, where a frontier reasoning row’s effective cost per hard problem stops being defensible. The trade is depth. It is the smaller model, so validate it against your hardest cases rather than your representative ones before moving a route onto it.
OpenAI o4-mini sits close enough to OpenAI o3 on price that the two are usually decided together. That comparison is written once, on the models table, rather than argued twice from opposite sides.
What OpenAI o4-mini costs in practice
A 50M/10M month is $99, which lands within a dollar of Claude Haiku 4.5 on identical volumes and with an identical 200K window — an unusually direct comparison between two vendors. Filling that window costs 22 cents of input. At $1.93 per million blended it is the seventh cheapest row in this table, and the caveat every reasoning row carries applies here too: reasoning tokens are billed as output, so the visible ratio understates a genuinely hard route.
| Scenario | Input / mo | Output / mo | Est. cost / mo |
|---|---|---|---|
| Prototype | 2M | 0.5M | $4 |
| Growing product | 50M | 10M | $99 |
| At scale | 500M | 100M | $990 |
Plug in your own numbers with the cost calculator.
Where OpenAI o4-mini fits in a routing setup
This is the row to start reasoning work on rather than the one to escalate to: run the reasoning-shaped routes here first and promote only the problems that measurably fail. Below it, DeepSeek-R1 does the reference month for about $49 against $99, if an open-weight reasoning model and an offshore API are both acceptable. Above it, the frontier reasoning row costs $180 for the same volumes, and considerably more once thinking tokens are counted rather than assumed.
Cheaper alternatives to OpenAI o4-mini
The models immediately below OpenAI o4-mini on blended cost, nearest first. A swap only pays if quality holds on the routes you move — see the LLM cost optimization playbook for how to prove that before you move them.
Frequently asked questions
How much does OpenAI o4-mini cost?
OpenAI o4-mini bills $1.10 per million input tokens and $4.40 per million output tokens — $1.93 per million blended at a three-to-one input:output split. A workload of 50M input and 10M output tokens per month works out at about $99. Confirm current pricing with the provider.
What is OpenAI o4-mini's context window?
OpenAI o4-mini has a 200K-token context window. One request that fills the whole window costs about $0.22 at the input rate, before the reply is billed.
Where does OpenAI o4-mini sit on price?
At $1.93 per million blended — three input tokens for every output token — OpenAI o4-mini is dearer than 6 of the 19 models tracked here and cheaper than 12. Output bills at 4x the input rate, and the nearest cheaper model on this table is DeepSeek-R1 at $0.96 blended.
Related concepts: context window, model routing, prompt caching, evals.