o4-mini
Fast o-series model for compact reasoning, coding, and tool use.
- Capability
- 145.2ECI · #91 of 158
- Input
- $1.10per 1M tokens
- Output
- $4.40per 1M tokens
- Context
- 200K100K max output
What it costs
- Input
- $1.10per million tokens
- Output
- $4.40per million tokens
- Cached input
- $0.275per million tokens
- Blended (3:1)
- $1.93pricier than 71% of priced models
Typical monthly bills
| Monthly usage | Estimated cost |
|---|---|
| Side project2M input + 0.5M output tokens | $4.40 |
| Team assistant25M input + 5M output tokens | $49.50 |
| Production app250M input + 50M output tokens | $495.00 |
Official OpenAI API price, as listed on models.dev.
Independent benchmarks
#91 of 158 scored models
The shaded band is Epoch AI’s confidence range (143.7–146.7); the tick marks the median scored model.
Scores from Epoch AI, run independently of OpenAI.
The details
- Lab
- OpenAI
- Released
- Apr 16, 2025
- Knowledge cutoff
- May 2024
- Context window
- 200,000 tokens
- Max output
- 100,000 tokens
- Inputs
- Text, Images
- Output
- Text
- Reasoning
- Adjustable effort low · medium · high
- Tool calling
- Yes
- Structured output
- Yes
- Weights
- Proprietary
- API model ID
o4-minion OpenAI- Availability
- 17 API providerslisted on models.dev
- Status
- Deprecated by the provider
What OpenAI claims
Published by the lab at launch. Settings vary, so compare these only with care.
| Benchmark | Score | Setting | Source |
|---|---|---|---|
| Aider Polyglot | 72 percent correct | — | Source |
More from OpenAI
About o4-mini
How much does o4-mini cost?
o4-mini costs $1.10 input / $4.40 output per million tokens (official OpenAI API price). Cached input is $0.275 per million tokens. At a 3:1 input-to-output mix that is $1.93 per million tokens, more expensive than 71% of the 469 priced models we track.
What is the context window of o4-mini?
o4-mini accepts up to 200,000 tokens per request and can write up to 100,000 tokens in one response.
How good is o4-mini?
Epoch AI gives o4-mini a Capabilities Index score of 145.2 (likely range 143.7–146.7), ranking it #91 of 158 models Epoch has scored. Epoch AI benchmark results: GPQA Diamond 79.6%, FrontierMath Tiers 1–3 36.1%, OTIS Mock AIME 2024–2025 81.7%, SimpleQA Verified 19.6%.
Is o4-mini open source?
No. o4-mini is proprietary; you use it through OpenAI’s API or partner platforms.
What inputs does o4-mini support?
o4-mini accepts text and images and replies in text. It is a reasoning model with low, medium and high effort settings, supports tool calling and can return structured JSON output.
When was o4-mini released?
OpenAI released o4-mini on Apr 16, 2025. Its training data runs to May 2024.
What are the best alternatives to o4-mini?
The closest current models from other labs on capability, price and release date are Gemini 2.5 Pro (Google, ECI 144.9, $1.25 / $10.00), Claude Haiku 4.5 (Anthropic, ECI 142.8, $1.00 / $5.00), Kimi K2 Thinking (Moonshot AI, ECI 145.5, $0.60 / $2.50) and Qwen3 235B-A22B Thinking 2507 (Alibaba (Qwen), ECI 143.3, $0.30 / $2.90).