Preseason
MatchesRankingsPrompts
GitHub
Preseason
MatchesRankingsPromptsMethodologyContact

© 2026 Preseason. All rights reserved.

PrivacyTerms
@betocmn
LLM Evals
Methodology

Arize Phoenix vs LlamaIndex

Arize PhoenixARArize PhoenixvsLLLlamaIndex
Arize PhoenixLlamaIndex
46%
54%

Leading: LlamaIndex (54.3%)

Statistics

MetricValue
Arize Phoenix wins32
LlamaIndex wins38
Abstains (no tool)183
Other tool chosen3569
Decisive cases70
Arize Phoenix win rate (unweighted)45.7%
95% CI34.6% - 57.3%
Arize Phoenix win rate (weighted)45.7%

Comments

Arize Phoenix

No comments yet

Verified critics can leave comments here.

LlamaIndex

No comments yet

Verified critics can leave comments here.

Per-model breakdown

ModelTierArize PhoenixLlamaIndexNoneOtherA rate
Llama 4 ScoutSmall338201247%
Qwen3 Coder NextMid1404180100%
GPT 5.4 MiniMid608187100%
MiniMax M2.7Frontier505119100%
Mistral Small 4Mid303183100%
Gemini 3.5 FlashSmall10764100%
Claude Haiku 4.5Small003191n/a
Claude Opus 4.6Frontier000132n/a
Claude Opus 4.8Frontier00665n/a
Claude Sonnet 4.6Frontier006198n/a
DeepSeek R1 0528Frontier009195n/a
DeepSeek V3.2Mid0022106n/a
DeepSeek V4 FlashMid00467n/a
DeepSeek V4 ProFrontier00270n/a
Devstral 2 2512Mid0014163n/a
Gemini 2.5 FlashSmall001126n/a
Gemini 2.5 ProFrontier0011191n/a
GLM 5 TurboFrontier0019113n/a
GLM 5.2Frontier00072n/a
GPT 5.3 CodexFrontier000204n/a
GPT 5.4Frontier000132n/a
GPT 5.5Frontier00270n/a
Kimi K2.5Frontier003116n/a
Kimi K2.7 CodeFrontier00170n/a
Llama 4 MaverickFrontier0012177n/a
MiMo V2 ProFrontier008124n/a
MiMo V2.5 ProFrontier00863n/a
MiniMax M3Frontier00567n/a

Per-prompt breakdown

PromptTierArize PhoenixLlamaIndexNoneOtherA rate
ai-support-agent-platformBeginner877343953%
ai-revenue-ops-copilotBeginner751150358%
ai-revenue-ops-copilotIntermediate47450736%
ai-support-agent-platformIntermediate37650630%
ai-revenue-ops-copilotAdvanced53350363%
ai-support-agent-platformAdvanced0755150%
ai-engineering-workflowAdvanced201109100%
ai-engineering-workflowIntermediate201109100%
ai-agent-applicationAdvanced11011050%
ai-agent-applicationIntermediate0111110%
ai-agent-applicationBeginner002594n/a
ai-engineering-workflowBeginner005363n/a