OpenAI · proprietary

GPT-5.5

GPT-5.5 by OpenAI appears in 3 sources with Reasoning at 62.25. Best read for Coding, High intelligence, LLM.

CreatorOpenAI

Release date2026-04-23

Knowledge cutoffNot published

Context1.1M tokens

Input price$5/M tokens

Output price$30/M tokens

Modalitytext + vision

CountryUS

Metrics

All source-backed metrics

Open in leaderboard

LLM Stats Rank3ranking · source_rank

Reasoning62.25reasoning · index_reasoning

Math48.43math · index_math

Coding51.04coding · index_code

Research36research · index_search

Writing30.76writing · index_communication

Vision44.98multimodal · index_vision

Tool calling38.33tool_calling · index_tool_calling

Long context29.44long_context · index_long_context

Finance30.73domain · index_finance

GPQA93.6 %reasoning · gpqa_score

Code Arena1,947.87 %coding · coding_arena_score

Humanity Last Exam52.2 %reasoning · hle_score

ARC-AGI v285 %reasoning · arc_agi_v2_score

MMMU-Pro83.2 %multimodal · mmmu_pro_score

BrowseComp84.4 %research · browsecomp_score

Toolathlon55.6 %tool_calling · toolathlon_score

MCP Atlas75.3 %tool_calling · mcp_atlas_score

FrontierMath35.4 %math · frontiermath_score

MRCR v274 %long_context · mrcr_v2_score

SWE-bench Pro58.6 %coding · swe_bench_pro_score

Context1,050,000 tokenscontext · context

Speed174.74 c/sperformance · throughput

Latency20,275.83 msperformance · latency

Input price5 $/Mpricing · input_price

Output price30 $/Mpricing · output_price

Arena Rating1,462.51metric

Arena Rank14metric

Vote Count8,686metric

GPQA Diamond93.6 %metric

Humanity's Last Exam41.4 %metric

Evidence

Citations and source overlap

Arena AILM Arena text leaderboard (overall) mirrored in the public leaderboard dataset.Retrieved 2026-05-20 VellumTop models per benchmark from page-level structured data (JSON-LD).Retrieved 2026-05-20 LLM StatsPublic full LLM leaderboard advanced explorer snapshot, including ranking, benchmark indexes, pricing, speed, context, license, and model metadata.Retrieved 2026-06-01

FAQ

How should I read this profile?

Treat this as a source-backed model dossier, not an EvalKit-run verification. The public values are replicated from linked sources and kept source-scoped.

Is GPT-5.5 verified by EvalKit?

No. EvalKit currently shows 0 verified rows until real run evidence exists.

Why can metrics disagree?

Different sources test different tasks, dates, prompts, and aggregation methods. EvalKit keeps those differences visible instead of merging them into a fake universal score.