← Model list

o4-mini

openai·openai/o4-mini-2·Deprecated·Closed·o-mini series

Fast o-series model for compact reasoning, coding, and tool use

At a glance
Official price$1.10 / $4.40 per 1M
Cache read $0.275 · OpenAI
Lowest paid$0.99 / $4.00
Poe Gateway · 1.1× spread
Context200,000
Output limit100,000
Capabilities
ReasoningTool useStructured outputTemperatureAttachments
⚠ Providers report capability flags inconsistently
Modalities
TextImagePDF
Knowledge cutoff2024-05
Released / updated2025-04-16 / 2025-04-16

Quality & performanceArtificial Analysis · Intelligence Index v4.1 · rep. tier high

Intelligence26.1
Value15
Coding
Agentic
Output speed143 tok/s
TTFT29.16 s
Value formulaIQ 26.1 ÷ min blended $1.742 = 15

Quality is independently evaluated by Artificial Analysis. Speed/latency are model-level medians.

Available at 17 providers15 with public prices

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
Poe
openai/o4-mini
Gateway$0.99$4.00$0.25200,000100,000
Impossibl
openai/o4-mini
Gateway$1.10$4.40$0.275200,000100,000
Jiekou.AI
o4-mini
Gateway$1.10$4.40200,000100,000
OpenRouter
openai/o4-mini
Gateway$1.10$4.40$0.275200,000100,000
OpenAIOfficial
o4-mini
First-party$1.10$4.40$0.275200,000100,000deprecated
NEAR AI Cloud
openai/o4-mini
Gateway$1.10$4.40$0.275200,000100,000
Merge Gateway
openai/o4-mini
Gateway$1.10$4.40$0.275200,000100,000
Cloudflare AI Gateway
openai/o4-mini
Cloud$1.10$4.40$0.275200,000100,000deprecated
Azure Cognitive Services
o4-mini
First-party$1.10$4.40$0.275200,000100,000deprecated
Azure
o4-mini
First-party$1.10$4.40$0.275200,000100,000deprecated

Sorted by blended price (input×0.75 + output×0.25) asc. The official channel always shows regardless of rank. Whether a gateway's low price is actually usable can't be verified.

Your usage cost

1Poe$309.20
2Impossibl$341.00
3OpenRouter$341.00
4OpenAI · Official$341.00
5NEAR AI Cloud$341.00
6Merge Gateway$341.00
Switch to Poe to save $31.80/mo (9%). The gap is small, so staying on the official channel is fine.
Note: this is a gateway; verify availability and rate limits yourself.

Benchmark1 items

NameConditionsScoreMetricSource
Aider Polyglot72percent correctSource ↗

The same benchmark scores very differently across harness / dataset, so the qualifying conditions must be shown together.

Reasoning control

effort = loweffort = mediumeffort = higheffort = none

2 / 17 providers expose no reasoning control (reasoning_options: []).

Related models

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$4.40$02026-08-052026-08-132026-08-05 · Input list $1.102026-08-06 · Input list $1.102026-08-07 · Input list $1.102026-08-08 · Input list $1.102026-08-09 · Input list $1.102026-08-10 · Input list $1.102026-08-11 · Input list $1.102026-08-12 · Input list $1.102026-08-13 · Input list $1.102026-08-05 · Output list $4.402026-08-06 · Output list $4.402026-08-07 · Output list $4.402026-08-08 · Output list $4.402026-08-09 · Output list $4.402026-08-10 · Output list $4.402026-08-11 · Output list $4.402026-08-12 · Output list $4.402026-08-13 · Output list $4.402026-08-05 · Min blended $1.932026-08-06 · Min blended $1.932026-08-07 · Min blended $1.932026-08-08 · Min blended $1.932026-08-09 · Min blended $1.932026-08-10 · Min blended $1.932026-08-11 · Min blended $1.932026-08-12 · Min blended $1.932026-08-13 · Min blended $1.74