Six RAG benchmarks. SynapseKit 1.4 vs LangChain 1.2 vs LlamaIndex 0.14. Kaggle CPU environment. All notebooks reproducible.
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
Week 2 Raw Values
2-Week Cumulative Standings
www.engineersofai.com · LLM Showdown · Notebooks #8–#14 · Kaggle CPU · All results reproducible