UCP PlaygroundGemini 2.5 Flash/Models/Gemini 2.5 Flash
InspectorAgentDocs
GoogleGoogle

Gemini 2.5 Flash

The fastest model in the benchmark at 11 seconds average. Moderate checkout rate but takes more turns (9.4 avg) than frontier models to get there. Low token cost makes it ideal for high-volume testing. Occasionally fabricates buyer data when identity linking is unavailable.

Avg Tokens44,478Avg Duration13.7sTurns to Checkout9.2
CheckoutCartSearchTurnsTokens25.738.973.33540
Shopping Score37/100
Fair

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%25.7%
Cart Rate20%38.9%
Search Rate10%73.3%
Turn Efficiency15%35%
Token Efficiency15%40%
25.7% of sessions failed with errors.

Top Stores

StoreCheckout %Cart %
••••••••••••••••••••••••••••••••100%100%
••••••••••••••••••••••••••••••••100%100%
thebodyshop.com100%100%
•••••••••••••••100%100%
perfume-empire.com66.7%66.7%
ucp.travel61.5%61.5%
kyliecosmetics.com57.1%57.1%
pier1.com50%50%
sini.fi50%50%
houseofparfum.nl49.2%61.9%

Known Issues

Data Fabricationmedium

Occasionally fabricates buyer data (fake names/emails) when identity linking is unavailable.

Turn Inefficiencylow

9.4 avg turns to checkout — takes more attempts than frontier models.

Token Usage

Avg per Session44,478
Fleet Average67,352
vs Fleet−34%
Median (p50)22,669
p90100,657
Prompt / Completion split
Prompt 43,465 (99%) Completion 367 (1%)
Daily Average Tokens (last 30 days)
Distribution range
Min0p259,693p5022,669p7545,179p90100,657Max860,031

Test Gemini 2.5 Flash on Your Store

Run a live agent session to see how Gemini 2.5 Flash handles your store's checkout flow end-to-end.

Run Agent Session

Compare With Other Models

MetaMetaLlama 3.3 70B39.3%GoogleGoogleGemini 3.5 Flash38.7%AnthropicAnthropicClaude Sonnet 4.636.7%xAIxAIGrok 4.326.3%GoogleGoogleGemini 2.5 Flash25.7%AnthropicAnthropicClaude Opus 4.823.3%DeepSeekDeepSeekDeepSeek V3.222.1%GoogleGoogleGemini 3.1 Pro21%GoogleGoogleGemini 2.5 Pro20%OpenAIOpenAIGPT-4o14.7%OpenAIOpenAIGPT-5.511.1%xAIxAIGrok 3 Mini (Reasoning)11.1%OpenAIOpenAIo4-mini (Reasoning)9.4%AnthropicAnthropicClaude Sonnet 58.9%DeepSeekDeepSeekDeepSeek V4 Pro8.6%xAIxAIGrok 4.57.7%AnthropicAnthropicClaude Opus 56.8%DeepSeekDeepSeekDeepSeek V4 Flash6.5%Moonshot AIMoonshot AIKimi K34.5%DeepSeekDeepSeekDeepSeek R1 (Reasoning)4.1%GoogleGoogleGemini 3.6 Flash3.2%MetaMetaLlama 4 Maverick2.6%AlibabaAlibabaQwen 3.7 Max0%AlibabaAlibabaQwQ 32B (Reasoning)0%
Full comparison view →