Model Arena

⚙️ [HARDWARE CONTROL & TELEMETRY]

Active VRAM utilization, queue depths, and Lineup states

Active Lineup & State

[actual_lineup]:
default_chat
[desired_lineup]:
default_chat
[hardware_lock]:
idle
Users: 1Media Queue: 0
Awaiting Valkey hardware SSE telemetry...

👔 [INTENT & ROUTING]

Execution trace and concurrent task blueprinting
Triage PhaseTracing...
Chief
Gatekeeper
Quartermaster

🗄️ [ASYNCHRONOUS GATHERING]

Multi-modal data extraction before optimization
Awaiting Orchestrator Dispatch...

🔬 [CONTEXT OPTIMIZATION]

Protecting the 32k context window and validating constraints
Awaiting Data Gathering...

🧠 [COGNITIVE EXECUTION]

Model performance stats and loaded configuration dossier
Context Window Map0 / 8,192 Tokens
System Rules RAG Payload

Live Inference Stats

TPS0.0
TTFT0ms
Tokens Generated0

Gemma-4-E2B-it (Base)

Agent Dossier
vLLM Configured
VRAM Target 5.5 GB
Quantization none
Base ModelAuto
Hostheads-core
Loading dossier...