Preseason
MatchesRankingsPrompts
GitHub
Preseason
MatchesRankingsPromptsMethodologyContact

© 2026 Preseason. All rights reserved.

PrivacyTerms
@betocmn
CMS
Methodology

DocFX vs Starlight

DODocFXvsSTStarlight
DocFXStarlight
44%
56%

Leading: Starlight (55.6%)

Insufficient data
This matchup has 9 decisive cases (minimum 30 required for publication).

Statistics

MetricValue
DocFX wins4
Starlight wins5
Abstains (no tool)197
Other tool chosen2967
Decisive cases9
DocFX win rate (unweighted)44.4%
95% CI18.9% - 73.3%
DocFX win rate (weighted)44.4%

Comments

DocFX

No comments yet

Verified critics can leave comments here.

Starlight

No comments yet

Verified critics can leave comments here.

Per-model breakdown

ModelTierDocFXStarlightNoneOtherA rate
Mistral Small 4Mid41015380%
DeepSeek V3.2Mid0361230%
MiniMax M2.7Frontier0151240%
Claude Haiku 4.5Small000168n/a
Claude Opus 4.6Frontier000132n/a
Claude Opus 4.8Frontier00036n/a
Claude Sonnet 4.6Frontier000168n/a
DeepSeek R1 0528Frontier0011156n/a
DeepSeek V4 FlashMid00134n/a
DeepSeek V4 ProFrontier00432n/a
Devstral 2 2512Mid000162n/a
Gemini 2.5 FlashSmall000132n/a
Gemini 2.5 ProFrontier002165n/a
Gemini 3.5 FlashSmall00531n/a
GLM 5 TurboFrontier004388n/a
GLM 5.2Frontier00036n/a
GPT 5.3 CodexFrontier006158n/a
GPT 5.4Frontier000132n/a
GPT 5.4 MiniMid007154n/a
GPT 5.5Frontier00036n/a
Kimi K2.5Frontier009121n/a
Kimi K2.7 CodeFrontier00431n/a
Llama 4 MaverickFrontier0026129n/a
Llama 4 ScoutSmall005158n/a
MiMo V2 ProFrontier004128n/a
MiMo V2.5 ProFrontier00134n/a
MiniMax M3Frontier00036n/a
Qwen3 Coder NextMid0058110n/a

Per-prompt breakdown

PromptTierDocFXStarlightNoneOtherA rate
documentation-siteIntermediate413048880%
documentation-siteAdvanced02145040%
documentation-siteBeginner021234040%
blog-platform-cmsAdvanced0030498n/a
blog-platform-cmsBeginner000537n/a
blog-platform-cmsIntermediate000536n/a