OpenAI · Frontier
GPT 5.5
GPT-5.5 is an OpenAI model, built for agentic coding, computer use, and professional knowledge work with stronger reasoning and token efficiency.
- Reasoning
- Tools
- Vision
- Structured output
- PDF input
- Input
- Text, Images
- Output
- Text
- Reasoning effort
- none · low · medium · high · xhigh
- Discovery tier
- Frontier
Pricing
USD per 1M tokens- Input
- $5.00
- Output
- $30.00
- Cache read
- $0.50
Reference prices as published by the model provider. API example
Specifications
- Context window
- 1.1M
- Max output
- 128K
- Average output (7 days)
- 490 tokens
- Longer than 48% of 473 models
Side by side
Compare models
Put GPT 5.5 next to up to three other models. The comparison stays in the link, so you can share it.
Public evaluations
Benchmarks
Independent evaluation results, with how each score ranks among the models in the Noumen catalog.
Intelligence Index
38.4
Tested at xhigh reasoning effort
Better than 82% of 326 scored models
Indexes
Coding Index
74.9Composite of the coding evaluations
Better than 89% of 168 scored models
Agentic work
AutomationBench-AA
47.3%Workflow automation, penalized for guardrail violations
Better than 61% of 173 scored models
AutomationBench-AA tasks completed
19.6%Workflows finished end to end without a guardrail violation
Better than 65% of 173 scored models
AA-Briefcasev1.1
1,136 EloAgentic knowledge work, rated head to head
Better than 56% of 166 scored models
GDPval-AAv2.1
1,351 EloEconomically valuable occupational tasks, rated head to head
Better than 61% of 187 scored models
Document reasoning
GDP.pdf
21.2%Professional reasoning over PDF documents, all-pass rate
Better than 78% of 171 scored models
AA-LCRv1.1
84.3%Reasoning across long documents
Better than 96% of 309 scored models
MLCR-AA
28.9%Long-context reasoning over medical records
Better than 90% of 121 scored models
Reasoning and knowledge
Humanity’s Last Exam
45.8%Expert-level questions across academic fields
Better than 88% of 323 scored models
GPQA Diamond
93.5%Graduate-level science questions
Better than 92% of 291 scored models
IFBench
75.9%Following precise output instructions
Better than 88% of 215 scored models
Coding
Terminal-Benchv4.0
14.6%Practical software engineering in a terminal
Better than 71% of 171 scored models
SciCode
55.8%Scientific computing in Python
Better than 78% of 174 scored models
Legacy evaluations
Terminal-Bench Hard
60.6%Earlier terminal task suite
Better than 98% of 212 scored models
τ²-Bench Telecom
93.9%Tool use in a customer-support simulation
Better than 81% of 214 scored models
AA-LCR (unversioned)
84.3%Long-context reasoning, earlier grading
Better than 96% of 309 scored models
Scores from Artificial Analysis, updated Oct 26, 2026. GPT 5.5 was tested at xhigh reasoning effort; compare scores measured at the same effort.Artificial Analysis
- #11Roleplay30.1%
- #15Math6.5%
Ranks among Noumen catalog models on Use cases leaderboards. Only top-20 placements are listed.
- #15Math1500.6
- #19Instruction Following1477.6
- #16German1490.3
- #10Industry Business And Management And Financial Operations1490.1
- #17Industry Legal And Government1495.1
- #15Industry Mathematical1506.3
- #8Japanese1505.1
- #16Korean1451.5
- #17Non English1473.6
- #10Spanish1485.0
- #13Vision Style Control Overall1286.0
- #13Vision Style Control · Creative Writing Vision1281.0
- #8Vision Style Control · Diagram1323.3
- #13Vision Style Control · English1280.9
- #14Vision Style Control · Entity Recognition1219.2
- #6Vision Style Control · Homework1335.1
- #8Vision Style Control · Humor1287.0
- #14Vision Style Control · Ocr1299.0
- #2Search Overall1246.6
- #6Document Overall1485.7
- #8Document Style Control Overall1482.8
Ranks among Noumen catalog models on LMArena leaderboards. Only top-20 placements are listed.LMArena
- #18FrontierCode Main43.0%
- #16FrontierCode Extended56.7%
Ranks among Noumen catalog models on FrontierCode leaderboards. Only top-20 placements are listed.FrontierCode
- #10Epoch Capabilities Index162.9
- #1FrontierMath v2 (Tiers 1–3)85.3%
- #1SWE-bench Verified80.6%
- #3GPQA Diamond94.0%
Ranks among Noumen catalog models on Epoch AI leaderboards. Only top-20 placements are listed.Epoch AI
- #12ARC-AGI-196.5%
- #11ARC-AGI-285.0%
- #14ARC-AGI-3 (Default harness)0.4%
Ranks among Noumen catalog models on ARC Prize leaderboards. Only top-20 placements are listed.ARC Prize
- #7EQ-Bench 41294
- #18Creative Writing v31845
- #8EQ-Bench 31554
Ranks among Noumen catalog models on EQ-Bench leaderboards. Only top-20 placements are listed.EQ-Bench
- #19SVG1233
- #15ASCII Art1251
- #12Agentic Game Dev1142
- #15Native Apps1165
Ranks among Noumen catalog models on Design Arena leaderboards. Only top-20 placements are listed.Design Arena
Live routing data
Performance and reliability
Recent averages across requests routed for this model. Your own speed depends on prompt length, reasoning effort and load.
- Prompt cache hits
- 0.0%
- About $5.00 per 1M input tokens with caching
- Request success (24h)
- 100.0%
- Reliable
Cache hits are the share of input tokens recently served from a prompt cache across all requests. They are not a forecast of your own hit rate.
- Healthy (98% or more)
- Degraded (90–98%)
- Down (under 90%)
- No requests
Developers
Use it through the API
Noumen’s API is OpenAI-compatible: any Chat Completions client works with a base URL and a key.
curl https://www.noumen.xyz/api/v1/chat/completions \
-H "Authorization: Bearer $NOUMEN_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "openai/gpt-5.5",
"reasoning_effort": "medium",
"messages": [{ "role": "user", "content": "Explain this stack trace." }]
}'import os
from openai import OpenAI
client = OpenAI(base_url="https://www.noumen.xyz/api/v1", api_key=os.environ["NOUMEN_API_KEY"])
reply = client.chat.completions.create(
model="openai/gpt-5.5",
messages=[{"role": "user", "content": "Explain this stack trace."}],
)
print(reply.choices[0].message.content)OpenAI
Related models
Other OpenAI models, starting with the same family.