InferX — Serverless GPU Inference Platform for Production Workloads

Tenant Namespace Model Name Type GPU Count vRam (GB) CPU Memory (GB) Standby State Snapshot Nodes Revision Actions
GPU Pageable Pinned
public Trial Huihui-Qwen3-Next-80B-A3B-Thinking-abliterated text2text 4 45.0 12.0 100.0 Mem File File Normal [] 255 Open
public Trial L3.3-70B-Loki-V2.0 text2text 2 71.0 20.0 100.0 Mem File File Normal ['computeinstance-e00r2jrqynf83a8b4f'] 259 Open
public Trial translategemma-27b-it-FP8-Dynamic text2text 1 32.0 20.0 80.0 Mem File File Fail [] 266 Open