Preseason
MatchesRankingsPrompts
GitHub
Preseason
MatchesRankingsPromptsMethodologyContact

© 2026 Preseason. All rights reserved.

PrivacyTerms
@betocmn
Agentic Web Search
Methodology

Tavily vs Serper

TATavilyvsSESerper
TavilySerper
89%
11%

Leading: Tavily (88.9%)

Statistics

MetricValue
Tavily wins40
Serper wins5
Abstains (no tool)0
Other tool chosen5
Decisive cases45
Tavily win rate (unweighted)88.9%
95% CI76.5% - 95.2%
Tavily win rate (weighted)88.9%

Comments

Tavily

No comments yet

Verified critics can leave comments here.

Serper

No comments yet

Verified critics can leave comments here.

Per-model breakdown

ModelTierTavilySerperNoneOtherA rate
Claude Opus 4.8Frontier3000100%
Claude Sonnet 4.6Frontier3000100%
DeepSeek V4 FlashMid3000100%
DeepSeek V4 ProFrontier3000100%
Gemini 3.5 FlashSmall3000100%
GLM 5.2Frontier3000100%
GPT 5.3 CodexFrontier3000100%
GPT 5.5Frontier3000100%
Kimi K2.7 CodeFrontier3000100%
MiMo V2.5 ProFrontier3000100%
MiniMax M3Frontier3000100%
GPT 5.4 MiniMid2001100%
Mistral Small 4Mid2001100%
Qwen3 Coder NextMid2000100%
DeepSeek R1 0528Frontier02010%
Llama 4 MaverickFrontier02000%
Claude Haiku 4.5Small1000100%
Llama 4 ScoutSmall01000%
Devstral 2 2512Mid0002n/a

Per-prompt breakdown

PromptTierTavilySerperNoneOtherA rate
ai-agent-applicationIntermediate1420188%
ai-agent-applicationBeginner1320287%
ai-agent-applicationAdvanced1310293%