# Auto-generated health check by SakThai cron job
eval_date: 2026-07-30T22:56:00Z
model: Nanthasit/sakthai-context-1.5b-merged-v2

metrics:
  downloads: 0
  likes: 0
  download_velocity_per_day: 0.0
  days_on_hub: 0.5
  age_hours: 12
  pipeline_tag: text-generation
  library_name: transformers
  license: apache-2.0
  base_model: Qwen/Qwen2.5-1.5B-Instruct
  last_modified: 2026-07-30T22:52:21Z
  created_at: 2026-07-30T10:46:30Z

storage:
  used_storage_bytes: 3098889036
  used_storage_human: "2955.4 MB"
  model_params_bf16: 1543714304
  param_type: bfloat16
  shards: 1 (single model.safetensors)
  total_siblings: 14
  hidden_files: 6
  non_hidden_files: 8

visibility:
  private: false
  gated: false

health_assessment: AMBER
health_score: 42
health_note: "Day-0 model (~12 hours old). 0 downloads expected. Good card and config but no benchmarks yet."
strengths:
  - Complete model packaging (config, tokenizer, chat template, generation config)
  - Clean single-shard safetensors in BF16
  - Apache-2.0 license
  - Well-documented base model and training datasets (v7, v8)
  - tag ecosystem complete (17 tags including tool-calling, conversational)
concerns:
  - 0 downloads, 0 likes (expected for day 0)
  - No published model-index/benchmarks
  - No GGUF variant for local inference
  - No eval-results data linked
recommendations:
  - Cross-link from higher-download siblings
  - Run tool-calling benchmarks and add model-index
  - Consider publishing GGUF variants
  - Promote via sakthai social channels