Gemini 3.7 Flash wins THOR Finding Triage Benchmark

zacharynado · x · 2026-08-15

Following the release of Gemini 3.7 Flash, tests on the THOR Finding Triage Benchmark show it has taken the #1 spot with a score of 72.5%, achieving 100% Threat Capture and 0% Critical Misses. Comparisons with Qwen 3.7 Max, Kimi K3, and DeepSeek V4 revealed that competitors generally struggled with false positives—escalating too many benign findings instead of filtering them out. Gemini 3.7 Flash excels precisely in catching threats while minimizing these false positives.

Original post →

More from Models

• Review: Running Qwen3.8-27B at 128K Context on a Single 32GB GPU — WSTangoDelta · 2026-08-15
• Paper reveals LM Head as gradient bottleneck amid Anthropic vocab size rumors — JFPuget · 2026-08-15
• Qwen3.8-27B Dual 3090 Benchmark: Detailed Perf Data and Configs — No-Statement-0001 · 2026-08-15
• Code World Model moves beyond static code prediction with internal world models — bendee983 · 2026-08-15
• Uncensored Qwen 3.8 27B 'Heretic' Released, Claimed Opus 4.6-Level — Temporary_Idea8880 · 2026-08-15
• Qwen 3.8 Now Available on Bittensor Subnet 95 — markjeffrey · 2026-08-15

Models channel →