Preseason
MatchesRankingsPrompts
GitHub
Preseason
MatchesRankingsPromptsMethodologyContact

© 2026 Preseason. All rights reserved.

PrivacyTerms
@betocmn
AI / LLM Integration
Methodology

Intercom vs Anthropic

INIntercomvsAnthropicANAnthropic
IntercomAnthropic
44%
56%

Leading: Anthropic (56.2%)

Statistics

MetricValue
Intercom wins130
Anthropic wins167
Abstains (no tool)128
Other tool chosen3563
Decisive cases297
Intercom win rate (unweighted)43.8%
95% CI38.2% - 49.5%
Intercom win rate (weighted)43.8%

Comments

Intercom

No comments yet

Verified critics can leave comments here.

Anthropic

No comments yet

Verified critics can leave comments here.

Model range breakdown

Model rangeTierIntercomAnthropicNoneOtherA rate
DeepSeek Flash 3.2–4.1Mid161918294%
DeepSeek Pro 4–4 0813Frontier00972n/a
DeepSeek R1Frontier1023186100%
DevstralMid2016167100%
Gemini Flash 2.5–3.8Small0612040%
Gemini Pro 2.5–3.1 PreviewFrontier21220767%
GLM 5 Turbo–5.3Frontier23320340%
GPT 5.4–6 AstraFrontier95019364%
GPT Coding 5.3 Codex–6.1 SolFrontier0302090%
GPT Mini 5.4 Mini–6 LunaMid91120090%
HaikuSmall99201119%
Kimi 2.5–3Frontier11131828%
Llama MaverickFrontier6027173100%
Llama ScoutSmall23011169100%
MiMo Pro 2–2.6Frontier42420067%
MiniMax M 2.7–3Frontier827016823%
Mistral SmallMid21111812466%
Opus 4.6–5.5Frontier000212n/a
Qwen CoderMid0212070%
Sonnet 4.6–5.5Frontier172019489%

Per-prompt breakdown

PromptTierIntercomAnthropicNoneOtherA rate
ai-support-agent-platformBeginner12912142591%
ai-revenue-ops-copilotIntermediate03415530%
ai-support-agent-platformIntermediate13105593%
ai-revenue-ops-copilotBeginner02215330%
ai-agent-applicationAdvanced019261300%
ai-agent-applicationBeginner015501130%
ai-agent-applicationIntermediate015441120%
ai-support-agent-platformAdvanced01415740%
ai-revenue-ops-copilotAdvanced0545640%