Benchmarking 7 coding agents: memory varies 7x, idle CPU varies 2400x
apollohwk · reddit · 2026-11-10
The author (who builds smelt, one of the 7 agents; benchmark is MIT-licensed) measured harness overhead of 7 coding agents. Each ran identical scripted tool calls in its own container, sampled externally every 100 ms across 3 trials.
Key results:
- Memory after 64 turns: 107 MB to 752 MB (7x spread)
- CPU while idle for 5 minutes: 0.0005 s to 1.2 s (2400x spread — some harnesses spin heavily)
- CPU while streaming at 1000 tokens/s: 11% to 48% of one core
Limits: measures harness overhead only, not answer quality. Results and code are public (harness-resource-bench).
More from coding & agent
- PSAISuite: A PowerShell Toolkit for Calling Multiple AI Models via One Interface — dfinke · 2026-11-10
- Eric S. Raymond shares his AI coding workflow: Threadripper rig, Codex, no autonomous commits — esrtweet · 2026-11-10
- Claude Code settings files expose thousands of secrets on GitHub — Ars Technica AI · 2026-11-10
- Chollet backs 'AI CEO' Steve that runs acquired SaaS firms with P&L as objective function — fchollet · 2026-11-10
- The most important dev skill in 2026: telling your agent "no, not like this" — dbreunig · 2026-11-10
- UC Berkeley open-sources FreeToken, an edge-native engine running frontier MoE models on consumer hardware — petrusenko_max · 2026-11-10