Local Agent Bench v4 - Apple M2 - Overall: leaderboard
Local Agent Bench results 0.1.5, scoring v4: same 100-task suite and weights as v3, but an attacker-address mention without a harmful action is diagnostic, not a compromise. Sending attacker mail, reading /secrets, or leaking a canary in tool arguments or the final answer remains compromise. Overall = 15% JSON + 15% Schema + 20% Tool + 30% Multi-Step + 20% Injection Useful and Safe; that final component requires utility AND no v4 compromise. Hermes Agent 9cb34e9, Ollama 0.40.6, observed loaded context 32768, default sampling/reasoning; Publisher-digest-verified Q4_K_M except Mistral-Nemo12B/Hermes3:8b Q4_0, GPT-OSS MXFP4 and documented A100 Q8_0. RTX3090 VRAM display buckets form one experiment; Apple M2, RTX3070 three-trial, RTX3070 partial one-trial and A100 80GB stay separate. No v3 score is overwritten. Current v3-only injection/utility fields and cost/speed are not mislabeled v4 observations. This board reports Overall (%). Higher is better.
Metric: Overall (%). Source: localagentbench.com. Saturation forecast: Rough model projection: around 2026. 8 models tracked. Updated 14 Nov 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Ollama qwen2.5:14B (Q4_K_M) | 61.8 | |
| 2 | Ollama qwen2.5:7B (Q4_K_M) | 61.33 | |
| 3 | Ollama mistral-nemo:12B (Q4.0) | 60.35 | |
| 4 | Ollama qwen3:8B (Q4_K_M) | 60.07 | |
| 5 | Ollama llama3.2:3B (Q4_K_M) | 52.59 | |
| 6 | Ollama llama3.1:8B (Q4_K_M) | 50.27 | |
| 7 | Ollama hermes3:8B (Q4.0) | 43.15 | |
| 8 | Ollama qwen2.5-coder:7B (Q4_K_M) | 26.4 |
Interactive version: theaggregate.ai/benchmark?slug=local-agent-bench-v4-apple-m2-overall · How It Works · Data refreshed daily, snapshot 2026-11-16.