← Model list

o3

openai·openai/o3-2·GA·Closed·o series

Deliberate o-series reasoner for hard math, coding, and multi-step analysis

At a glance
Official price$2.00 / $8.00 per 1M
Cache read $0.50 · OpenAI
Lowest paid$1.80 / $7.20
Poe Gateway · 5.6× spread
Context200,000
⚠ Providers report 131,072–200,000; the table below is authoritative
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

Intelligence31.1
Value10
Coding
Agentic
Output speed112 tok/s
TTFT8.31 s
Value formulaIQ 31.1 ÷ min blended $3.150 = 10

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

Available at 16 providers15 with public prices

ProviderTierInputOutputCache readCache writeContextOutput limitStatus
Poe
openai/o3
Gateway$1.80$7.20$0.45200,000100,000
Impossibl
openai/o3
Gateway$2.00$8.00$0.50200,000100,000
OpenRouter
openai/o3
Gateway$2.00$8.00$0.50200,000100,000
OpenAIOfficial
o3
First-party$2.00$8.00$0.50200,000100,000
NEAR AI Cloud
openai/o3
Gateway$2.00$8.00$0.50200,000100,000
Merge Gateway
openai/o3
Gateway$2.00$8.00$0.50200,000100,000
Cloudflare AI Gateway
openai/o3
Cloud$2.00$8.00$0.50200,000100,000
Azure Cognitive Services
o3
First-party$2.00$8.00$0.50200,000100,000
Azure
o3
First-party$2.00$8.00$0.50200,000100,000
Vercel AI Gateway
openai/o3
Cloud$2.00$8.00$0.50200,000100,000

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$558.00
2Impossibl$620.00
3OpenRouter$620.00
4OpenAI · Official$620.00
5NEAR AI Cloud$620.00
6Merge Gateway$620.00
Switch to Poe to save $62.00/mo (10%). 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 Polyglot81.3percent 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 / 16 providers expose no reasoning control (reasoning_options: []).

Related models

Price historyone sample accumulated per data sync

Input listOutput listMin blended
$8.00$02026-08-052026-08-132026-08-05 · Input list $2.002026-08-06 · Input list $2.002026-08-07 · Input list $2.002026-08-08 · Input list $2.002026-08-09 · Input list $2.002026-08-10 · Input list $2.002026-08-11 · Input list $2.002026-08-12 · Input list $2.002026-08-13 · Input list $2.002026-08-05 · Output list $8.002026-08-06 · Output list $8.002026-08-07 · Output list $8.002026-08-08 · Output list $8.002026-08-09 · Output list $8.002026-08-10 · Output list $8.002026-08-11 · Output list $8.002026-08-12 · Output list $8.002026-08-13 · Output list $8.002026-08-05 · Min blended $3.502026-08-06 · Min blended $3.502026-08-07 · Min blended $3.502026-08-08 · Min blended $3.502026-08-09 · Min blended $3.502026-08-10 · Min blended $3.502026-08-11 · Min blended $3.502026-08-12 · Min blended $3.502026-08-13 · Min blended $3.15