Cad Bench — Development skill for Claude Code
Benchmark latency, throughput, and memory/VRAM footprint before vs. after.
How to install Cad Bench
Installs to ~/.claude/skills/sayanthegamer-cadence-cad-bench/SKILL.md
mkdir -p ~/.claude/skills/sayanthegamer-cadence-cad-bench && curl -fsSL https://raw.githubusercontent.com/Sayanthegamer/cadence/HEAD/commands/cad-bench.md -o ~/.claude/skills/sayanthegamer-cadence-cad-bench/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Cad Bench does
name: cad-bench description: Benchmark latency, throughput, and memory/VRAM footprint before vs. after.
Execute the `cadence-bench` skill:
- Target the performance-critical function or loop.
- Measure baseline metrics (latency in ms, peak memory in MB, throughput in ops/sec) over warm iterations.
- Apply code optimization or changes.
- Run comparative benchmark with identical inputs and seeds.
- Emit the Efficiency Diff table (flagging any regression >5%).
Alternatives in Development
- Investigate Performance — Investigate a Nimbalyst performance problem (freeze, lag, idle CPU, slow op, memory growth) with measured evid 1.6k ★
- Projectmem — Open-source coding agent memory 766 ★
- Mem Save — Save a curated session-level summary to history 186 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Blast Radius Bench
A benchmark for agentic coding-tool judgment under ambiguity: does the agent confirm before touching ambiguous
Claw Bench
Benchmark suite: Claw Code (Rust) vs Claude Code (Node.js)
Warden Bench
Run the golden benchmark suite for one token-warden agent (or all) and compare results against the frozen run1
Gemma4 Turboquant Bench
Gemma 4 26B + TurboQuant: 262K context on a single RTX 4090. Agentic coding benchmark via Claude Code.
Harness Bench
Benchmark harness for Codex, Claude Code, and Cursor debugging runs
Run Rolling Aggregate Bench
Run the IVM benchmark from the rolling-aggregate-bench/ directory.
Related Agents
Bench
API performance benchmarking — latency profiling, throughput testing, performance regression detection
Perf Tester
Use to measure latency, throughput, and load behaviour of HTTP endpoints or hot functions with autocannon or k
Bench Reporter
Updates benchmark documentation with results from bench-runner. Maintains BENCHMARK.md, README.md, and EVAL-GA