UCP PlaygroundGPT-5.5/Models/GPT-5.5
InspectorAgentDocs
OpenAIOpenAI

GPT-5.5

OpenAI's frontier model (GPT-5.5, succeeding 5.2). The prior 5.2 revision had a notably high tool-error rate in our benchmark — nearly half of tool calls failed against MCP schemas, though turn efficiency was reasonable (~7.6 turns) when it reached checkout. 5.5 is the newer revision and is being re-benchmarked.

Avg Tokens59,438Avg Duration33.9sTurns to Checkout10.1
CheckoutCartSearchTurnsTokens10.224.570.2248
Shopping Score21/100
Weak

Scores computed from real agent sessions against live UCP-enabled stores. Not estimated — every data point is from an actual tool call, checkout attempt, and store response.

Shopping Score Breakdown

Checkout Rate40%10.2%
Cart Rate20%24.5%
Search Rate10%70.2%
Turn Efficiency15%24%
Token Efficiency15%8%
29.8% of sessions failed with errors.

Top Stores

StoreCheckout %Cart %
••••••••••••••••100%100%
••••••••••••••••85.7%85.7%
••••••••••••••••50%50%
kyliecosmetics.com50%50%
••••••••••••••••37.5%37.5%
••••••••••••••••33.3%33.3%
houseofparfum.nl30.4%39.1%
apneeswimwear.com25%25%
everlane.com22.2%22.2%
ucp.travel20%20%

Known Issues

High Tool Error Ratehigh

46.9% of all tool calls fail. Nearly half of MCP interactions produce errors.

Schema Compliancehigh

Struggles with MCP tool schemas — incorrect argument types, missing required fields.

Token Usage

Avg per Session59,438
Fleet Average75,080
vs Fleet−21%
Median (p50)24,259
p90132,727
Prompt / Completion split
Prompt 58,280 (98%) Completion 1,158 (2%)
Daily Average Tokens (last 30 days)
Distribution range
Min0p2510,628p5024,259p7556,171p90132,727Max1,427,124

Test GPT-5.5 on Your Store

Run a live agent session to see how GPT-5.5 handles your store's checkout flow end-to-end.

Run Agent Session

Compare With Other Models

MetaMetaLlama 3.3 70B39.3%GoogleGoogleGemini 3.5 Flash38.1%AnthropicAnthropicClaude Sonnet 4.634.7%xAIxAIGrok 4.326.3%GoogleGoogleGemini 2.5 Flash23.3%AnthropicAnthropicClaude Opus 4.822.5%AlibabaAlibabaQwen 3.8 Flash22.2%DeepSeekDeepSeekDeepSeek V3.220.7%GoogleGoogleGemini 2.5 Pro19.8%GoogleGoogleGemini 3.1 Pro19.7%OpenAIOpenAIGPT-4o15%DeepSeekDeepSeekDeepSeek V4.1 Flash14.3%xAIxAIGrok 3 Mini (Reasoning)11.1%OpenAIOpenAIGPT-5.510.2%AnthropicAnthropicClaude Opus 59.7%OpenAIOpenAIo4-mini (Reasoning)9.1%xAIxAIGrok 4.58.1%AnthropicAnthropicClaude Sonnet 58.1%DeepSeekDeepSeekDeepSeek V4 Pro7.7%DeepSeekDeepSeekDeepSeek V4 Flash7.5%GoogleGoogleGemini 3.8 Flash6.7%DeepSeekDeepSeekDeepSeek R1 (Reasoning)3.6%Moonshot AIMoonshot AIKimi K33.1%AlibabaAlibabaQwen 3.7 Max2.6%MetaMetaLlama 4 Maverick2.3%GoogleGoogleGemini 3.6 Flash1.1%GoogleGoogleGemini 3.7 Flash0%OpenAIOpenAIGPT-6 Luna0%MetaMetaMuse Spark 1.30%AlibabaAlibabaQwQ 32B (Reasoning)0%
Full comparison view →