GPT-5.5 (medium)
OpenAIGPT
Description
GPT-5.5 is OpenAI's smartest model yet, designed for real work across agentic coding, computer use, knowledge work, and early scientific research. It matches GPT-5.4 per-token latency in real-world serving while reaching a much higher level of intelligence and using significantly fewer tokens to complete the same tasks. GPT-5.5 supports a 1M-token context window in the API and a 400K-token context window in Codex, with state-of-the-art results on Terminal-Bench 2.0, OSWorld-Verified, GDPval, FrontierMath, and CyberGym.
Release Date
2026-04-23
Parameters
—
Context Length
1.1M
Modalities
image, pdf, text
Capability Radar
48
general
69
coding
93
reasoning
69
scienceest.
80
agents
85
multimodal
Science uses a reasoning proxy when dedicated science benchmarks are unavailable.
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Code Ranking | 4 | 94.0 | AA |
| General Ranking | 8 | 86.0 | AA |
| Science | 8 | 88.0 | AA |
Benchmark Scores (LLM Stats)
Agents
GDPval-AA
1135.00 / 3000SR
BrowseComp
84.4%SR
Terminal-Bench 2.0
82.7%SR
CyberGym
81.8%SR
BixBench
80.5%SR
OSWorld-Verified
78.7%SR
MCP Atlas
75.3%SR
FrontierSWE
73.0%SR
Finance Agent
60.0%SR
SWE-Bench Pro
58.6%SR
Toolathlon
55.6%SR
OfficeQA Pro
54.1%SR
Finance Agent v2
51.8%SR
GeneBench
25.0%SR
Legal Agent Benchmark
2.1%SR
Biology
GPQA
93.6%SR
Communication
Tau2 Telecom
98.0%SR
Finance
GDPval-MM
84.9%SR
General
MMMU-Pro
83.2%SR
LiveBench
80.7%SR
MRCR v2 (8-needle)
74.0%SR
Long Context
Graphwalks parents >128k
58.5%SR
Graphwalks BFS >128k
45.4%SR
Math
Humanity's Last Exam
52.2%SR
FrontierMath
35.4%SR
Reasoning
ARC-AGI
95.0%SR
ARC-AGI v2
85.0%SR
AA Evaluation Indices
Coding Index71.5
Intelligence Index50.4
Gpqa0.9
Tau20.9
Terminalbench V2 10.8
Lcr0.7
Ifbench0.7
Terminalbench Hard0.6
Scicode0.5
Hle0.4
Tau Banking0.3
LLM Stats Category Scores
Legal100
Finance100
General100
Agents88
Reasoning52
Communication100
Physics90
Biology90
Chemistry90
Multimodal80
Safety80
Search80
Tool Calling80
Vision80
Spatial Reasoning70
Code70
Long Context60
Math60
Pricing
Input Price$5 / 1M tokens
Output Price$30 / 1M tokens
Blended Price (3:1)$11.25 / 1M tokens
Cache Read Price$0.5 / 1M tokens
Speed
Tokens/sec70.3
Time to First Token6.92s
Time to Answer6.92s
Provider Price Ranking
Provider Price Ranking
1 providers
ProviderInputOutput
1OpenAIPRIMARY
$5
$30
Compare pricing across different API providers for this model.