Automated page. Written and translated automatically by AI and checked against the sources by software; no human editor. SI-IQ is an index built from public benchmark results, not an IQ test. How it works

How smart is GPT-5.5?

138 SI-IQ · 90% interval 135–141
Rank 9 of 80 ranked models · Plausible rank range 3–13

GPT-5.5 is a model from OpenAI, released on Apr 23, 2026. Its SI-IQ is 138 (90% interval 135–141), estimated from 9 benchmark results by 2 independent evaluators. Its strongest area is Reasoning and its weakest is Math.

Facts

Developer: OpenAI
Released: Apr 23, 2026
Access: API access
Benchmarks used: 9 results from 2 evaluators
API price: $5.00 input / $30 output per 1M tokens (developer's price list, Oct 18, 2026)
Context window: 1,050,000 tokens

Ability by area (SI-IQ, 90% interval)

Every benchmark result we used

Each score is one value exactly as the source published it: for Epoch AI the best of its run settings, for OpenRouter the median hosting provider (the developer's own hosting when it ran the test). “Implied SI-IQ” is the ability that this single result points to; the overall SI-IQ is their precision-weighted average.

Benchmark	Area	Score	Implied SI-IQ	Evaluated by	As listed by the source	Retrieved	Source
Chess Puzzles	Reasoning	63.0%	141	Epoch AI	`gpt-5.5_xhigh`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
Furniture Assembly not in index	Reasoning	44.2%	—	Epoch AI	`gpt-5.5_xhigh`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
Mystery Game Puzzles	Reasoning	56.0%	140	Epoch AI	`gpt-5.5_xhigh`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
FrontierMath Tier 4	Math	72.5%	137	Epoch AI	`gpt-5.5_xhigh`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
FrontierMath Tiers 1–3	Math	85.3%	137	Epoch AI	`gpt-5.5_xhigh`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
OTIS Mock AIME 2024–2025	Math	100.0%	142	Epoch AI	`gpt-5.5-pre-release_xhigh`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
SWE-bench Verified (Epoch AI run)	Coding	80.6%	132	Epoch AI	`gpt-5.5-pre-release_xhigh`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
GPQA Diamond	Knowledge	94.0%	135	Epoch AI	`gpt-5.5-pre-release_xhigh`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
GPQA Diamond (OpenRouter run)	Knowledge	93.6%	141	OpenRouter	`openai/gpt-5.5-20260423` OpenAI · first-party provider (1 of 3 rows)	Oct 18, 2026	OpenRouter benchmark datasets · data file
SimpleQA Verified	Knowledge	63.0%	142	Epoch AI	`gpt-5.5_xhigh`	Oct 18, 2026	Epoch AI Benchmarking Hub · data file
τ²-bench Airline (OpenRouter run)	Agents	75.0%	129	OpenRouter	`openai/gpt-5.5-20260423` OpenAI · first-party provider (1 of 3 rows)	Oct 18, 2026	OpenRouter benchmark datasets · data file

Human preference (Arena)

On the Arena leaderboard, where people vote between two anonymous answers, GPT-5.5 has a rating of 1,477 (1,473–1,480) from 70,985 votes (published Oct 15, 2026). Human preference is a different measure from benchmark ability and is not part of SI-IQ. Arena (Hugging Face)

Models with similar SI-IQ

Model	SI-IQ
Claude Fable 5Anthropic	138	GPT-5.5 vs Claude Fable 5
Claude Opus 5Anthropic	136	GPT-5.5 vs Claude Opus 5
Claude Fable 5.1Anthropic	140	GPT-5.5 vs Claude Fable 5.1
GPT-5.6 SolOpenAI	140	GPT-5.5 vs GPT-5.6 Sol
Gemini 4 ArgonGoogle DeepMind	140	GPT-5.5 vs Gemini 4 Argon

Computed with SI-IQ method version 1.0.0. Read the full methodology

Data sources: Arena leaderboard dataset on Hugging Face (CC BY 4.0); Epoch AI, ‘Capabilities & benchmarking’ (CC BY 4.0); API list prices via LiteLLM's price table (MIT); each price links to the developer's pricing page; Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0.