Local Agent Bench - RTX3090 - Injection Resistance: leaderboard
Local Agent Bench RTX3090 population: Injection Resistance on a public deterministic 100-task tool-use suite. Hermes Agent/Ollama default sampling and reasoning, loaded 32K context, exact GGUF/quant artifacts; timeouts count as failures. Scoring v3 distinguishes useful-and-safe completion from attack resistance. Memory display tiers are not separate scored experiments.
Metric: Injection Resistance (%). Source: localagentbench.com. Saturation forecast: Rough model projection: around 2026. 27 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Ollama gemma4:e4b (Local Agent Bench Run: quant=Q4_K_M;gguf=370c2879f17648d337b0ab4e208fdce348e24d1060064e93d7645daafd56d39c;engine=ollama0.40.6;reasoning=default;trials=3) | 100 | |
| 2 | Ollama qwen2.5-coder:7b (Local Agent Bench Run: quant=Q4_K_M;gguf=60e05f2100071479f596b964f89f510f057ce397ea22f2833a0cfe029bfc2463;engine=ollama0.40.6;reasoning=default;trials=3) | 100 | |
| 3 | Ollama gemma4:12b (Local Agent Bench Run: quant=Q4_K_M;gguf=bb722270d54346adc198f213851dbe0207e87c3ecf2a0aff4d92262726215391;engine=ollama0.40.6;reasoning=default;trials=3) | 100 | |
| 4 | Ollama qwen2.5:14b (Local Agent Bench Run: quant=Q4_K_M;gguf=2049f5674b1e92b4464e5729975c9689fcfbf0b0e4443ccf10b5339f370f9a54;engine=ollama0.40.6;reasoning=default;trials=3) | 100 | |
| 5 | Ollama deepseek-r1:8b (Local Agent Bench Run: quant=Q4_K_M;gguf=e6a7edc1a4d7d9b2de136a221a57336b76316cfe53a252aeba814496c5ae439d;engine=ollama0.40.6;reasoning=default;trials=3) | 100 | |
| 6 | Ollama deepseek-r1:14b (Local Agent Bench Run: quant=Q4_K_M;gguf=6e9f90f02bb3b39b59e81916e8cfce9deb45aeaeb9a54a5be4414486b907dc1e;engine=ollama0.40.6;reasoning=default;trials=3) | 100 | |
| 7 | Ollama qwen2.5-coder:14b (Local Agent Bench Run: quant=Q4_K_M;gguf=ac9bc7a69dab38da1c790838955f1293420b55ab555ef6b4615efa1c1507b1ed;engine=ollama0.40.6;reasoning=default;trials=3) | 100 | |
| 8 | Ollama mistral:7b (Local Agent Bench Run: quant=Q4_K_M;gguf=f5074b1221da0f5a2910d33b642efa5b9eb58cfdddca1c79e16d7ad28aa2b31f;engine=ollama0.40.6;reasoning=default;trials=3) | 100 | |
| 9 | Ollama gemma4:26b (Local Agent Bench Run: quant=Q4_K_M;gguf=a8bdd07a49b895083bb1420bb895ad9ab3e3b8f7d5cc42c7582ca0fa765eede2;engine=ollama0.40.6;reasoning=default;trials=3) | 100 | |
| 10 | Ollama qwen3.6:27b (Local Agent Bench Run: quant=Q4_K_M;gguf=415aad607fc297a2ad84cda238febbde295f49f17c17ad7784e2246060fb06d0;engine=ollama0.40.6;reasoning=default;trials=3) | 100 |
Interactive version: theaggregate.ai/benchmark?slug=local-agent-bench-rtx3090-injection-resistance · How It Works · Data refreshed daily, snapshot 2026-11-05.