All times Eastern (ET)Provider pages checked daily at
ProSign inRSS

o4 mini (2025-04-16)

API ID o4-mini-2025-04-16

o4-mini-2025-04-16 is an OpenAI model, released April 2025. It has a 200,000-token context window and up to 100,000 output tokens. It takes image input and supports function calling, structured outputs and reasoning. Standard price: from $1.10 input and from $4.40 output per 1M tokens.

Status
Deprecated
Released
April 2025
Shutdown
Oct 23, 2026

It stops serving on October 23, 2026. Recommended replacement: gpt-5.6-terra.

How to migrate to GPT-5.6 Terra

Free tier: Free for shared data, 10M tokens a day for Launch and Grow (2.5M for Build), in place when tracking began (October 12, 2026) (source) · All free models

Benchmarks

Scores as each source published them
or

Comparing with o3 mini (2025-01-31), the previous model in this line. A score is compared only with the same benchmark from the same source, at the same setting where both have it. Full comparison: prices, limits, features

Arena

Arenao4 mini (2025-04-16)o3 mini (2025-01-31)ChangeRankDateSource
Text
By category
  • Math 1418 rank 132 vs 1382 (+36)
  • Creative writing 1337 rank 193 vs 1301 (+36)
  • Instruction following 1368 rank 188 vs 1343 (+25)
  • Hard prompts 1406 rank 181 vs 1371 (+35)
  • Long prompts 1369 rank 202 vs 1359 (+10, within range)
  • Multi-turn 1383 rank 186 vs 1340 (+43)
  • Expert 1407 rank 167 vs 1365 (+42)
1391 ±41348 ±4+43173 of 415
44,856 votes		Oct 13, 2026source
Text: Coding1433 ±71416 ±6+17178 of 410
8,656 votes		Oct 13, 2026source
Vision1201 ±7not rated90 of 165
41,341 votes		Oct 9, 2026source

Rated as o4-mini-2025-04-16; o3 mini (2025-01-31) as o3-mini. Ratings come from people voting between two anonymous answers; Text and Vision are style-controlled, as on arena.ai. Data: Arena (arena.ai), CC BY 4.0.

Epoch AI

Benchmarko4 mini (2025-04-16)o3 mini (2025-01-31)ChangeSettingRunSource
FrontierMath Tier 44.9%0.0%+4.9 ptshigh effortJun 11, 2026source
FrontierMath Tiers 1-3
Other settings
  • 28.8% vs 10.5% medium effort
  • 16.1% vs 3.9% low effort
36.1%18.6%+17.5 ptshigh effortJun 11, 2026source
GPQA Diamond
Other settings
  • 77.8% vs 74.3% medium effort
  • 75.3% vs 68.2% low effort
79.6%77.0%+2.6 ptshigh effortApr 16, 2025source
MATH Level 597.8%96.5%+1.3 ptshigh effortApr 16, 2025source
OTIS Mock AIME 2024-2025
Other settings
  • 73.3% vs 63.9% medium effort
  • 57.8% vs 44.4% low effort
81.7%76.9%+4.8 ptshigh effortApr 16, 2025source
SimpleQA Verified
Other settings
  • 18.8% vs 15.3% high effort
19.6%15.3%+4.3 ptslow effort (vs high effort)Aug 27, 2026source

Run independently by Epoch AI. Data: Epoch AI, Capabilities & benchmarking (epoch.ai/benchmarks), CC BY.

On cloud platforms

All platform dates
PlatformPlatform IDStateShutdownTime leftReplacementSource
Microsoft Foundryversion 2025-04-16o4-miniDeprecatedNov 19, 202635 daysGPT-5.6 Terragpt-5.6-terrasource

Microsoft Foundry (version 2025-04-16) retires this model 27 days after OpenAI's own API.

Read daily from each platform's own pages. The shutdown date and countdown on this page use OpenAI's own date. Time left counts to 00:00 UTC on the shutdown date, which is the evening before in ET.

Aliases pointing here: o4-mini (in place when tracking began (October 7, 2026))

Features

Compare models
FeatureValueSinceSource
Structured outputsThe model can be held to a JSON schema you supply, so every reply parses without retries or cleanup.Supportedin place when tracking began (October 7, 2026)source
Function callingAlso called tool use. You describe functions; the model replies with a function name and arguments for your code to run, then uses the result.Supportedin place when tracking began (October 7, 2026)source
Prompt cachingA prompt prefix you send again and again (system prompt, documents, tools) is cached, so later requests cost less and start faster.Supportedin place when tracking began (October 7, 2026)source
Batch APISend many requests as one job that runs in the background, usually finished within 24 hours, at a lower price than live requests.Supportedin place when tracking began (October 7, 2026)source
Image inputImages such as screenshots, photos and charts can go in the prompt for the model to read.Supportedin place when tracking began (October 7, 2026)source
Reasoning / thinkingThe model can work through a problem step by step before it answers: usually better answers to hard questions, at the cost of more tokens and time.Supportedin place when tracking began (October 7, 2026)source
Fine-tuningYou can train your own version of the model on your examples through the company's API, then call it like any other model.Supportedin place when tracking began (October 7, 2026)source
Context windowThe most the model can work with in one request, in tokens: your prompt, conversation history and files. A token is about three-quarters of an English word.200,000 tokensin place when tracking began (October 7, 2026)source
Max output tokensThe longest reply the model can write in one response, in tokens. A token is about three-quarters of an English word.100,000 tokensin place when tracking began (October 7, 2026)source

Not published by OpenAI: PDF input, Built-in web search, temperature parameter, top_p parameter, logprobs.

Price history

USD per 1M tokens, standard tier
$0.00$1.00$2.00$3.00$4.00$5.00Oct 2026In $1.10Out $4.40Oct 1, 2026 to Oct 15, 2026: Input $1.10, Output $4.40
$0.00$1.00$2.00$3.00$4.00$5.00Oct 2026In $1.10Out $4.40Oct 1, 2026 to Oct 15, 2026: Input $1.10, Output $4.40

We started recording this model's prices on Oct 1, 2026. The history grows with each daily check.

Show as a table
FromTypePrice
Oct 1, 2026Input$1.10
Oct 1, 2026Output$4.40

Prices

USD per 1M tokens, every type and band, by tier
TypeBandPrice $/1MIn effect
Standard
inputshort context$1.10
cached inputshort context$0.275
outputshort context$4.40
Batch
inputshort context$0.55
outputshort context$2.20
Flex
inputshort context$0.55
cached inputshort context$0.138
outputshort context$2.20
Fast
inputshort context$2.00
cached inputshort context$0.50
outputshort context$8.00

Changelog

Changes we detected on OpenAI's pages

No changes recorded yet.

Time until shutdown
7days
11hours
25min
21sec

Counting to 8:00 pm ET on October 22, 2026, which is 00:00 UTC on October 23, 2026.

Email alerts

Get an email before o4-mini-2025-04-16 shuts down

We'll email you 30, 7 and 1 days before it shuts down, and the same morning if the shutdown date moves or the price changes.

No password: one email to confirm, and you can unsubscribe any time. Privacy

Last checked · Sources: OpenAI deprecations, OpenAI pricing, OpenAI models