Lilith Version 1 - tests and benchmarks

Automated design tests validate chat history storage and optional live Ollama sessions for Version 1. Cards below show recorded console benchmarks with execution time and model metadata.

Bench machine

Current system specs

CPUAMD Ryzen 5 4500 6-Core Processor

GPUNVIDIA GeForce RTX 3060

RAM32 GB

OSWindows 11 (10.0.26200)

Ollama127.0.0.1:11434

Recorded2026-05-21 11:10

Validation

Recorded console sessions

Benchmarks run against a local Ollama endpoint on the machine described in Test specs. Use these transcripts to compare latency and behavior before upgrading to a release with TTS or tool calling.

automated · 71.92s

Chat and history — gemma4

Model gemma4Time 71.92s

chat-and-history automated lilith run on v1 (gemma4). Total session time 71.92s.

View recording & transcript

automated · 11.79s

Chat and history — gemma4

Model gemma4Time 11.79s

chat-and-history automated lilith run on v1 (gemma4). Total session time 11.79s.

View recording & transcript

automated · 65.59s

Chat and history — llama3.2

Model llama3.2Time 65.59s

chat-and-history automated lilith run on v1 (llama3.2). Total session time 65.59s.

View recording & transcript