Week 2 Scorecard — Full Evidence Dashboard

Six RAG benchmarks. SynapseKit 1.4 vs LangChain 1.2 vs LlamaIndex 0.14. Kaggle CPU environment. All notebooks reproducible.

Week 2 Winner
SK
SynapseKit — 15.0 pts
Week 2 Runner-up
LI
LlamaIndex — 11.5 pts
Cumulative Leader
30
SynapseKit after 2 weeks

Points Breakdown — All 6 Benchmarks

Benchmark SynapseKit LlamaIndex LangChain
#8 RAG from PDF LoC — lower better 3.0 2.0 1.0
#9 Chunking Strategies Splitter count — higher better 1.0 3.0 2.0
#10 Built-in BM25 Extra packages — lower better 3.0 1.5 1.5
#11 Hybrid Search RRF Configurability /5 — higher better 2.0 1.0 3.0
#12 Streaming TTFT ms overhead — lower better 3.0 2.0 1.0
#13 Conversation Memory LoC to add memory — lower better 3.0 2.0 1.0
TOTAL 15.0 #1 11.5 #2 9.5 #3

Total Points — Stacked by Benchmark

SynapseKit
LlamaIndex
LangChain

Week 2 Raw Values

2-Week Cumulative Standings

www.engineersofai.com · LLM Showdown · Notebooks #8–#14 · Kaggle CPU · All results reproducible