OpenAI · model passport

OpenAI: GPT-5.1

gpt-5-1

Cloud only
Proprietary licence
400,000 context
Released 2025-11-13(from model name)

Overall · LMArena: 1,456#59 of 396
Best percentile: 97.3LiveCodeBench (Artificial Analysis)
Cohort standing: 64.7median 53.6

What the data says

Benchmarked on 22 of 27 tracked axes. Every figure below is the model's own published result on that board.

Strongest axes

97.3 pct — LiveCodeBench (Artificial Analysis) (86.8%)
96.6 pct — MMLU-Pro (Artificial Analysis) (87.0%)
90 pct — LiveBench (73.9%)

Weakest axes

5.3 pct — Document Arena (1,416)

Uneven: 92 percentile points between its best and worst axis — 97.3 pct on LiveCodeBench (Artificial Analysis) versus 5.3 pct on Document Arena.

LiveCodeBench (Artificial Analysis)97.3 pct86.8%Verify →
MMLU-Pro (Artificial Analysis)96.6 pct87.0%Verify →
LiveBench90 pct73.9%Verify →
IFBench (Artificial Analysis)87.7 pct72.9%Verify →
Instruction following86.1 pct1,450Verify →
Overall85.4 pct1,456Verify →
Hard prompts84.6 pct1,477Verify →
Coding79 pct1,492Verify →
FrontierMath75.4 pct31.0%Verify →
Factuality Arena74.4 pct1,458Verify →
GPQA Diamond74.2 pct87.6%Verify →
Vision Arena71.3 pct1,251Verify →
AIME (math)68.5 pct88.6%Verify →
Humanity's Last Exam63 pct23.7%Verify →
SimpleQA Verified62.2 pct48.0%Verify →
Terminal-Bench61.4 pct47.6%Verify →
SimpleBench57.5 pct53.2%Verify →
ARC-AGI-249.4 pct17.6%Verify →
WebDev Arena37.4 pct1,396Verify →
HHEM Hallucination (Vectara)35.2 pct89.1%Verify →
SWE-bench Verified25 pct68.0%Verify →
Document Arena5.3 pct1,416Verify →
Search ArenaNot benchmarked
Agent ArenaNot benchmarked
BFCL v4 (function calling)Not 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-5.1

Context length400,000 tokens

Max output128,000 tokens

Acceptsimage, text, file

Input price$1.25 per 1M tokens

Cached input price$0.13 per 1M tokens

Output price$10 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-5.1","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-5.1-high stands at 64.7 (raw mean 66.7) — measured on 22 of 27 axes.
standingtemplateVerify →
gpt-5.1-high stands at 64.7 versus a cohort median of 53.6 across 132 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 →