o3 Pro
Jun 2025
o3 ProCompare inon
Benchmarks of o3 Pro9
BenchmarkByScoreRank
Overall1
Reasoning4
Agents2
Legacy · saturated or no longer reported2
No benchmark matches
Rankingin thinking mode?
Strong at
Alternatives?
ModelScoreCostPer taskWhat changes
CheaperClaude Opus 5.5max×2.6-77%$5.98No forced tool use+5BetterGPT-5.6 Lunamedium+14%-99%$0.016
OpenMiMo v2.6 Prothinking×2.1-98%$0.13Pass back reasoning_content+3CheaperClaude Sonnet 5.5max×2.6-89%$5.46No forced tool use+5CheaperClaude Fable 5.1max×2.4-43%$7.63Replace forced tool_choice+5ModelScoreCostPer taskWhat changes
Claude Opus 5.5max×2.6-77%$5.98No forced tool use+5Claude Sonnet 5.5max×2.6-89%$5.46No forced tool use+5Claude Fable 5.1max×2.4-43%$7.63Replace forced tool_choice+5GPT-6 Astramax×2.4-43%$3.26Monitoring can return HTTP 403+2
Gemini 4 Argonhigh×2.4-89%*$1.99Synchronous output stops at 65,536+2ModelScoreCostPer taskWhat changes
GPT-5.6 Lunamedium+14%-99%$0.016GPT-6 Lunamedium+37%-99%$0.017Function calling only with none · Chat Completions+3Claude Haiku 5.5low+35%-99%$0.024Remove sampling parameters+5MiMo v2.5 ProthinkingOpen+19%-98%$0.054No image inputMiMo v2.6 FlashthinkingOpen+73%-100%$0.062Pass back reasoning_content+3ModelScoreCostPer taskWhat changes
MiMo v2.6 ProthinkingOpen×2.1-98%$0.13Pass back reasoning_content+3
DeepSeek V4.1 Flash 1017maxOpen×2.1-99%$0.26Pass back reasoning_content · Chat Completions+4
GLM 5.3maxOpen×2-94%$2.01No image input+4
Step 5 PreviewthinkingOpen+100%-96%$1.03web_search not available · Chat Completions+2
Kimi K3maxOpen+99%-83%$2.00Sampling parameters are fixed+3ModelScoreCostPer taskWhat changes
GPT-6 Astramedium×2.3—$1.54Monitoring can return HTTP 403+2GPT-6.1 Solmedium×2.2—$0.21Remove unsupported parameters+5GPT-6 Solmedium+82%—$0.25GPT-5.6 Solmedium+79%—$0.50GPT-5.5medium+55%—$0.90Prompts with >272K input tokensGPT-5.3 Codexxhigh+49%—$1.75 / $14DeprecatedGPT-5.6 Terramedium+38%—$0.18Prompts with >272K input tokensGPT-6 Lunamedium+37%—$0.017Function calling only with none · Chat Completions+3All OpenAI models →History
2 versionsFAQ
How much does o3 Pro cost?›
$20 per 1M input tokens and $80 per 1M output tokens.
Is o3 Pro free to use?›
There is no free API; the cheapest paid price is $20 per 1M input tokens.
How good is o3 Pro, and what is it best at?›
#105 of 334 in Chat & Knowledge, #2 of 15 in Coding, #15 of 16 in Agents & Tool Use.
Is o3 Pro better than Kimi K2 Thinking?›
On AA Intelligence Index 4.3.2, o3 Pro scores 21.9 vs 22 — about the same.
How can I use o3 Pro?›
Through the API of OpenAI.
What is the context window of o3 Pro?›
200K tokens, with up to 100K output tokens.
When was o3 Pro released? Is there a newer version?›
Released 2025-06-10. It has been superseded by GPT-5.
Is o3 Pro open source?›
No, o3 Pro is a closed model available through an API. The closest open-weights option here is MiMo v2.6 Pro.
Can o3 Pro read images?›
It accepts text and images, PDF files as input.
What is o3 Pro?›
o3 Pro is a closed language model from OpenAI, released Jun 2025.
What are the best alternatives to o3 Pro?›
Claude Opus 5.5 (cheaper), GPT-5.6 Luna (better), MiMo v2.6 Pro (open weights).