Navigate
AI BENCHY
Advertise here

Anthropic: Claude Sonnet 5 vs Google: Gemini 3.5 Flash-Lite

Gemini 3.5 Flash-Lite (medium) leads on average score with 6.5 vs 6.3. Gemini 3.5 Flash-Lite (medium) has the lower benchmark cost at $0.369 vs $0.548. Gemini 3.5 Flash-Lite (medium) is faster at 6.01s vs 6.04s, with pass rates of 45.5% vs 69.7%.

Recommended modelGemini 3.5 Flash-Lite (medium)It has the strongest score in this comparison (6.5) and the best overall balance of cost and response time across all 2 models.

Last updated at: 2026-07-21

Metric Claude Sonnet 5 Claude Sonnet 5 none Release: 2026-06-30 Gemini 3.5 Flash-Lite Gemini 3.5 Flash-Lite medium Release: 2026-07-21
Score 6.3 6.5
Rank #118 #104
Reliability 10.0 10.0
Consistency 8.7 7.4
Tests Correct
Attempt pass rate 45.5% 69.7%
Flaky tests 4 7
Total Runs 66 66
Cost per result 6.840 3.074
Total Cost $0.548 $0.369
Input Price $2.000 / 1M $0.300 / 1M
Output Price $10.000 / 1M $2.500 / 1M
Total Input Tokens 161,035 118,818
Output Tokens 22,511 11,677
Reasoning Tokens 0 121,611
Response Time (avg) 6.04s 6.01s
Response Time (max) 33.39s 49.03s
Response Time (total) 132.85s 132.30s

Hamster playing table tennis

Prompt: Create a detailed SVG illustration of a hamster playing table tennis.

#118 Claude Sonnet 5

none
Cost
$0.061
Time
53.7s
Tokens
6,172 tok

#104 Gemini 3.5 Flash-Lite

medium
Cost
$0.015
Time
16.4s
Tokens
5,971 tok

Top Models by Score

Score vs Total Cost

Response Time (avg)

Score vs Response Time (avg)

Total Output Tokens

Score vs Total Output Tokens

Category Breakdown

Coding Score Consistency Attempt pass rate Flaky tests Tests Correct Response Time (avg) Input Tokens Output Tokens Reasoning Tokens
Claude Sonnet 5 4.6 7.9 22.2% 1 3.67s 10,590 1,864 0
Gemini 3.5 Flash-Lite 5.5 7.4 44.4% 1 8.37s 8,136 452 25,999

Quick Compare

Switch Comparison Pair