o1
The original reasoning model: thinks before it answers.
About
o1, OpenAI's first "o-series" reasoning model (preview September 12, 2024; full release December 5, 2024), changed the scaling recipe: instead of bigger models, it spends more compute at inference time — thinking through hidden chain-of-thought steps before answering. That made it dramatically better than GPT-4o at math (AIME), coding (Codeforces), and graduate-level science (GPQA). 200K-token context, text+image input, proprietary/closed. An o1-pro tier ($200/mo ChatGPT Pro, $150/$600 per 1M via API) allocates even more reasoning compute. Superseded as the flagship by o3, but historically pivotal.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported (OpenAI): 74.4% AIME 2024 (o1-preview), 78% o1 / 86% o1-pro on AIME 2024; strong Codeforces/GPQA Diamond vs GPT-4o — provider-reported. Independent analyses confirmed the reasoning gains while noting higher cost/latency.
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 7/10
Scored 7/10 (verified tier, metrics-v1). Evidence: provider-tier estimate — thinner evidence (no public adoption metric); released 677 days ago. Manual review: the reasoning paradigm shift (RL on CoT, Dec 2024, 200K); superseded by o3; expensive ($15/$60 per 1M); reasoning tokens inflate cost/latency. License: Proprietary — closed weights.
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 9.0/10
- Maintenance · recency, releases, issue load
- 5.0/10
- Openness · license + docs
- 4.4/10
- Manual review · sanity read + red-flag check
- 8.0/10
- · provider-tier estimate — thinner evidence (no public adoption metric)
- · released 677 days ago
- · manual review: the reasoning paradigm shift (RL on CoT, Dec 2024, 200K); superseded by o3; expensive ($15/$60 per 1M); reasoning tokens inflate cost/latency