OpenAI · model passport

OpenAI: GPT-5.2

gpt-5-2

Cloud only
Proprietary licence
400,000 context
Released 2025-12-11(from model name)

Overall · LMArena: 1,437#91 of 396
Best percentile: 98.9LiveCodeBench (Artificial Analysis)
Cohort standing: 68.0median 54.2

What the data says

Benchmarked on 23 of 27 tracked axes. Every figure below is the model's own published result on that board.

Strongest axes

98.9 pct — LiveCodeBench (Artificial Analysis) (89.4%)
97.8 pct — MMLU-Pro (Artificial Analysis) (87.4%)
91.4 pct — LiveBench (74.6%)

Weakest axes

10.5 pct — Document Arena (1,425)

Uneven: 88.4 percentile points between its best and worst axis — 98.9 pct on LiveCodeBench (Artificial Analysis) versus 10.5 pct on Document Arena.

LiveCodeBench (Artificial Analysis)98.9 pct89.4%Verify →
MMLU-Pro (Artificial Analysis)97.8 pct87.4%Verify →
LiveBench91.4 pct74.6%Verify →
IFBench (Artificial Analysis)91.4 pct75.4%Verify →
FrontierMath91.3 pct40.7%Verify →
GPQA Diamond87.8 pct91.4%Verify →
BFCL v4 (function calling)Dormant board84.1 pct55.9%Verify →
AIME (math)83 pct96.1%Verify →
Terminal-Bench81.8 pct64.9%Verify →
Coding78 pct1,489Verify →
Hard prompts78 pct1,461Verify →
Instruction following77.8 pct1,427Verify →
Overall77.3 pct1,437Verify →
Humanity's Last Exam71.7 pct27.8%Verify →
Vision Arena64.7 pct1,243Verify →
ARC-AGI-261.8 pct52.9%Verify →
HHEM Hallucination (Vectara)60 pct91.6%Verify →
Factuality Arena54.1 pct1,442Verify →
SWE-bench Verified50 pct73.8%Verify →
WebDev Arena43.9 pct1,417Verify →
SimpleBench41.4 pct45.8%Verify →
SimpleQA Verified40.2 pct37.1%Verify →
Document Arena10.5 pct1,425Verify →
Search ArenaNot benchmarked
Agent ArenaNot benchmarked
SWE-bench Verified (live board)Not benchmarked
Terminal-Bench 4.0Not benchmarked

Percentiles are the model's standing within each board's own population. Axes marked "not benchmarked" have no published result for this model — that is an absence of data, not a low score.

How to run it

Use it hosted

API model IDopenai/gpt-5.2

Context length400,000 tokens

Max output128,000 tokens

Acceptsfile, image, text

Input price$1.75 per 1M tokens

Cached input price$0.17 per 1M tokens

Output price$14 per 1M tokens

API parameters this endpoint supports

Include reasoningMax completion tokensMax tokensReasoningReasoning effortResponse formatSeedStructured outputsTool choiceTool callingVerbosity

Example request

curl https://openrouter.ai/api/v1/chat/completions \
  -H "Authorization: Bearer $OPENROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"openai/gpt-5.2","messages":[{"role":"user","content":"Hello"}]}'

View on OpenRouter →

Identifiers, limits and prices are synced from the upstream source shown above. Any field we haven't synced is left out entirely rather than guessed.

gpt-5.2-high stands at 68 (raw mean 70.3) — measured on 23 of 27 axes.
standingtemplateVerify →
gpt-5.2-high stands at 68 versus a cohort median of 54.2 across 123 models released within 180 days — above the peer median.
cohorttemplateVerify →

Every claim is a computed fact from the current board data, phrased by a low-temperature model call and rejected by a validation gate if any number is not in the source row. How the leaderboards work →