OpenAIproprietary
Released: 2024-12-05

o1

OpenAI full-scale reasoning model using extensive chain-of-thought verification before responding.

Intelligence Index
87.2/ 100
Calibrated multi-domain composite
Throughput Speed
45tok/sec
Output streaming throughput
Time To First Token
950ms
Initial chunk server latency
LiveBench Contam-Free
73.5%
September 2026 suite score

LiveBench Multi-Domain Evaluation Breakdown

Monthly refreshed contamination-free test sets
Logical Reasoning86.8%
Coding & Repo Repair74.0%
Mathematics (AIME)89.2%
Data Analysis & Tables77.0%
Language & Comprehension82.0%
Instruction Following87.5%

Sourced Benchmark Evaluations (5)

Verified performance across authoritative benchmarks with provenance tracking.
BenchmarkStatusScoreTrustDateSource TypeProvenance
Humanity's Last ExamActive9.1%
90
2025-01-24independentSource ↗
ARC-AGI-2Nearing Saturation67.7%
75
2025-01-02independentSource ↗
SWE-bench VerifiedSaturated56.2%
25
2024-12-14independentSource ↗
IFEvalDeprecated87.5%
25
2024-12-09independentSource ↗
GPQANearing Saturation78%
75
2024-09-12vendor-reportedSource ↗
OpenAI Compatible API ExecutionAPI Ready
from openai import OpenAI

client = OpenAI()
response = client.chat.completions.create(
    model="o1",
    messages=[{"role": "user", "content": "Evaluate multi-step logic problem"}],
    temperature=0.2
)
print(response.choices[0].message.content)

Evaluation & Sourcing Notes

Scores listed for o1 represent verified evaluations extracted from official research papers, independent evaluation suites (HELM, LMSYS, OpenCompass, LiveBench), and verified audit reports.

All benchmarks marked as saturated or deprecated reflect historical performance where the benchmark no longer provides active discriminative power.