o4-mini
OpenAI o4-mini is a compact reasoning model in the o-series, optimized for fast, cost-efficient performance while retaining strong multimodal and agentic capabilities.
o4-miniReleased 16 Apr 2025 (Apr 2025)Pricing
List prices at OpenAI, in US dollars per million tokens.
| Tier | Input / 1M | Output / 1M | |
|---|---|---|---|
| Standard | $1.10 | $4.40 | |
| Cached input 75% off input | $0.275 | 75% off input | |
| Batch 50% off | $0.55 | $2.20 | 50% off |
| Flex 50% off | $0.55 | $2.20 | 50% off |
| Priority 82% more | $2.00 | $8.00 | 82% more |
| Web search $10.00 per 1,000 searches | $10.00 per 1,000 searches |
What it costs a month
At OpenAI list prices, 30 days.
- Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$586
- RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$488
- Coding agent1,000 requests a day, 40K in / 2K out, cached$891
- Overnight batch20,000 requests a day, 4K in / 0.4K out, batch$1,848
Where to run it
4 hosts serve o4-mini. The comparison is on blended price, 3 parts input to 1 part output.
| Host | In / out per 1M | Model id | |
|---|---|---|---|
OpenAILab API o4-mini | $1.10 / $4.40 | o4-mini | |
ReplicateInference o4-mini | $1.00 / $4.00 −9% vs headline | o4-mini | |
Azure OpenAICloud o4-mini | $1.10 / $4.40 Retires 19 Nov 2026 | o4-mini | Retires 19 Nov 2026 |
Vercel AI GatewayGateway o4-mini | $1.10 / $4.40 | o4-mini |
Show 7 more listings: regions, snapshots and variantsHide regional listings
| OpenAI | |||||||
| 2025-04-16 o4-mini-2025-04-16 | $1.10 / $4.40 | o4-mini-2025-04-16 | |||||
| 2025-04-16 · fine-tuning ft:o4-mini-2025-04-16 | $4.00 / $16.00 | ft:o4-mini-2025-04-16 | |||||
| Azure OpenAI | |||||||
| 2025-04-16 o4-mini-2025-04-16 | $1.10 / $4.40 Retires 19 Nov 2026 | o4-mini-2025-04-16 | Retires 19 Nov 2026 | ||||
| us o4-mini | $1.21 / $4.84 +10% vs headline Retires 19 Nov 2026 | o4-mini | Retires 19 Nov 2026 | ||||
| eu o4-mini | $1.21 / $4.84 +10% vs headline Retires 19 Nov 2026 | o4-mini | Retires 19 Nov 2026 | ||||
| us · 2025-04-16 o4-mini-2025-04-16 | $1.21 / $4.84 +10% vs headline Retires 19 Nov 2026 | o4-mini-2025-04-16 | Retires 19 Nov 2026 | ||||
| eu · 2025-04-16 o4-mini-2025-04-16 | $1.21 / $4.84 +10% vs headline Retires 19 Nov 2026 | o4-mini-2025-04-16 | Retires 19 Nov 2026 | ||||
Capabilities
Core
- Streaming
- Function calling
- JSON mode
- Structured outputs
Multimodal
- Vision
- Image input
- Audio
- Image generation
- Video
Advanced
- Reasoning
- Tool calling
- MCP
- Prompt caching
- Batch API
- Embeddings
- Takes in
- TextImagesPDFs
- Produces
- Text
- Features
- PDF inputPrompt cachingBuilt-in web search
Cheaper alternatives
Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.
Same model, cheaper host
- Replicate9% lesso4-mini$1.00 / $4.00
Smaller OpenAI models
- 12% less
- 76% less
- 90% less
Other frontier labs
- 22% less
- 35% less
- 90% less
- 56% less
Open-weight picks
- 10% less
- 64% less
- 73% less
- 75% less
Benchmarks
Scores published on the model card.
- Aider Polyglot72.0%
Model line
Newer and older versions, at each lab's own price.
Call it
Using OpenAI's API.
from openai import OpenAI
client = OpenAI() # reads OPENAI_API_KEY
response = client.responses.create(model="o4-mini", input="Hello")
print(response.output_text)curl https://api.openai.com/v1/responses \
-H "Authorization: Bearer $OPENAI_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "o4-mini", "input": "Hello"}'from litellm import completion
response = completion(
model="openai/o4-mini",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)Quick answers
How much does o4-mini cost?
On OpenAI, o4-mini costs $1.10 per million input tokens and $4.40 per million output tokens. Cached input costs $0.275 per million tokens. The cheapest host is Replicate, at $1.00 per million input tokens and $4.00 per million output tokens.
Which APIs offer o4-mini?
o4-mini runs on 4 APIs: OpenAI, Replicate, Azure OpenAI and Vercel AI Gateway.
What is o4-mini's context window?
200K tokens, with up to 100K tokens in one reply.
Is o4-mini being retired?
Azure OpenAI retires it on 19 Nov 2026. 3 hosts keep serving it.
What are cheaper alternatives to o4-mini?
Models that cost less and keep most of what it can do: GPT-5.4 Mini on OpenAI (12% less), Gemini 3.8 Flash on Google AI Studio (22% less), Grok Build 0.1 on xAI (35% less) and Inkling on DeepInfra (10% less). o4-mini itself costs 9% less on Replicate. Savings are on a mix of 3 input tokens to 1 output token.
When was o4-mini released?
OpenAI released it on 16 Apr 2025.
What is o4-mini's API model id?
Use o4-mini with OpenAI.
Sources
Raw catalog entry for o4-mini on OpenAI
{
"cache_read_input_token_cost": 2.75e-7,
"cache_read_input_token_cost_flex": 1.38e-7,
"cache_read_input_token_cost_priority": 5e-7,
"deprecation_date": "2026-10-23",
"deprecation_source": "https://developers.openai.com/api/docs/deprecations",
"successor_model": "gpt-5.6-terra",
"input_cost_per_token": 0.0000011,
"input_cost_per_token_batches": 5.5e-7,
"input_cost_per_token_flex": 5.5e-7,
"input_cost_per_token_priority": 0.000002,
"litellm_provider": "openai",
"max_input_tokens": 200000,
"max_output_tokens": 100000,
"mode": "chat",
"output_cost_per_token": 0.0000044,
"fedramp_uplift_multiplier": 1.1,
"output_cost_per_token_batches": 0.0000022,
"output_cost_per_token_flex": 0.0000022,
"output_cost_per_token_priority": 0.000008,
"search_context_cost_per_query": {
"search_context_size_high": 0.01,
"search_context_size_low": 0.01,
"search_context_size_medium": 0.01
},
"source": "https://developers.openai.com/api/docs/pricing",
"supports_function_calling": true,
"supports_parallel_function_calling": false,
"supports_pdf_input": true,
"supports_prompt_caching": true,
"supports_reasoning": true,
"supports_response_schema": true,
"supports_tool_choice": true,
"supports_vision": true,
"supports_web_search": true
}