Skip to content
aiqbench

OpenAI · model passport

OpenAI: GPT-5.5

gpt-5-5
  • Cloud only
  • Proprietary licence
  • 1,050,000 context
  • Released 2026-09-13(first seen on LMArena)
18/27axes covered
Overall · LMArena
1,481#23 of 396
Best percentile
100.0Terminal-Bench
Cohort standing
80.2median 60.3

What the data says

Benchmarked on 18 of 27 tracked axes. Every figure below is the model's own published result on that board.

Strongest axes

  • 100 pct — Terminal-Bench (84.7%)
  • 99.7 pct — LiveBench (79.9%)
  • 96.5 pct — Factuality Arena (1,486)

Uneven: 43.8 percentile points between its best and worst axis — 100 pct on Terminal-Bench versus 56.2 pct on HHEM Hallucination (Vectara).

  • Terminal-Bench100 pct84.7%Verify →
  • LiveBench99.7 pct79.9%Verify →
  • Factuality Arena96.5 pct1,486Verify →
  • Instruction following95.5 pct1,478Verify →
  • Overall94.4 pct1,481Verify →
  • Hard prompts93.4 pct1,501Verify →
  • IFBench (Artificial Analysis)92.9 pct75.9%Verify →
  • Coding91.8 pct1,519Verify →
  • ARC-AGI-289.9 pct85.0%Verify →
  • Vision Arena89.3 pct1,286Verify →
  • SimpleBench85.1 pct69.0%Verify →
  • GPQA Diamond84.2 pct90.7%Verify →
  • SimpleQA Verified84.1 pct63.0%Verify →
  • Document Arena81.6 pct1,483Verify →
  • Agent Arena70.9 pct0.033Verify →
  • WebDev Arena65.5 pct1,513Verify →
  • AIME (math)62.5 pct84.4%Verify →
  • HHEM Hallucination (Vectara)56.2 pct90.7%Verify →
  • Search ArenaNot benchmarked
  • FrontierMathNot benchmarked
  • Humanity's Last ExamNot benchmarked
  • BFCL v4 (function calling)Not benchmarked
  • SWE-bench Verified (live board)Not benchmarked
  • Terminal-Bench 4.0Not benchmarked
  • MMLU-Pro (Artificial Analysis)Not benchmarked
  • LiveCodeBench (Artificial Analysis)Not benchmarked
  • SWE-bench VerifiedNot 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-5.5
Context length1,050,000 tokens
Max output128,000 tokens
Acceptsfile, image, text
Input price$5 per 1M tokens
Cached input price$0.50 per 1M tokens
Output price$30 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-5.5","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-5.5-high stands at 80.2 (raw mean 85.2) — measured on 18 of 27 axes.

    standingtemplateVerify →

  • gpt-5.5-high stands at 80.2 versus a cohort median of 60.3 across 176 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 →