LLM.Kosol.net Intelligent Service

High-Performance Secure AI Inference Engine

ALL SYSTEMS OPERATIONAL
⏱️ Uptime LIVE
1h 13m
⚡ Active Latency NORMAL
1619 ms
🚦 Active / Queued GPU QUEUE
0 active / 0 queued
📍 Client IP CALLER
216.73.216.216
⚡ Active In-Memory Models (VRAM / GPU Status)
⚡ 1 Loaded in VRAM
🤖 qwen3.8:27b
unsloth/Qwen3.8-27B-GGUF
⚡ 100% GPU
💾 VRAM: Active in VRAM 100.0% Loaded on GPU
📐 27B⏳ TTL: Active (llama.cpp)🎯 Context: 81,920 (80k)
Requests: 0 CPU Spills: 0
Last Used: 2026-09-15 17:21:55 First Used: 2026-09-15 17:21:55
📦 Available AI Models (Installed)
6 Models Ready Unsloth Desktop (llama.cpp)
🤖 qwen3.8:27b
unsloth/Qwen3.8-27B-GGUF
⚡ In VRAM
📐 27B
🤖 gemma4:12b
unsloth/gemma-4-12B-it-qat-GGUF
📐 12B
🤖 gemma4:26b
unsloth/gemma-4-26B-A4B-it-GGUF
📐 26B
🤖 gemma4:e4b
unsloth/gemma-4-E4B-it-qat-GGUF
📐 E4B
🤖 unsloth/Qwen3.5-4B-MTP-GGUF
📐 4B
🤖 unsloth/Qwen3.5-9B-GGUF
📐 9B