# Inference Check - 20260731T001536Z
# Model: Nanthasit/sakthai-context-1.5b-merged-v2

inference_api:
  url: https://api-inference.huggingface.co/models/Nanthasit/sakthai-context-1.5b-merged-v2
  dns_resolution: false
  dns_error: "[Errno -5] No address associated with hostname"

inference_router_hf_inference:
  url: https://router.huggingface.co/hf-inference/models/Nanthasit/sakthai-context-1.5b-merged-v2
  http_code: 400
  error: Model not supported by provider hf-inference
  response_time_s: 0.144

local_inference:
  status: OOM
  available_ram_mb: 898
  model_size_estimate_fp16_gb: 3
  model_size_estimate_4bit_mb: 900
  root_cause: Insufficient RAM for 1.5B model loading

system_info:
  total_ram_mb: 7940
  free_ram_mb: 898
  swap_mb: 0
  python: 3.13.5
  torch: 2.13.0
  transformers: 5.14.1

hf_hub_info:
  model_exists: true
  pipeline_tag: text-generation
  library_name: transformers
  private: false
  base_model: Qwen/Qwen2.5-1.5B-Instruct
  downloads: 0
  inference_provider_mapping: null

verdict: FAIL - Inference API unreachable from cron environment (DNS) and local OOM