{
  "measured_at": "2026-09-02T06:36:22Z",
  "label": "quality-tier-gte-qwen2-q4km",
  "model": "hf.co/second-state/gte-Qwen2-1.5B-instruct-GGUF:Q4_K_M",
  "dimensions": 1536,
  "host": {
    "cpus": 8,
    "note": "CPU-only; assert size_vram == 0 in /api/ps before trusting these"
  },
  "fixture": {
    "corpus_id": "upsquad-platform-v1",
    "documents": 40,
    "document_chars_median": 378.0,
    "queries": 20,
    "query_chars_median": 86.5
  },
  "footprint": {
    "disk_bytes": 1116696312,
    "resident_bytes": 1566709416
  },
  "bulk_ingest": {
    "scope": "one request carrying all documents; the ingest leg only",
    "rounds": [
      {
        "seconds": 20.222090653143823,
        "embeddings_per_second": 1.9780348474395455
      },
      {
        "seconds": 19.704443921800703,
        "embeddings_per_second": 2.029998926066855
      },
      {
        "seconds": 19.695564183872193,
        "embeddings_per_second": 2.0309141503422476
      }
    ],
    "embeddings_per_second_median": 2.029998926066855,
    "embeddings_per_second_min": 1.9780348474395455,
    "embeddings_per_second_max": 2.0309141503422476
  },
  "query_path": {
    "scope": "the EMBEDDING leg of a search only \u2014 excludes ANN, BM25, RRF and recency fusion; this is the part that changes with the model",
    "samples": 20,
    "ms_p50": 171.33889393880963,
    "ms_p95": 235.40418967604637,
    "ms_max": 257.82263930886984,
    "ms_min": 126.22943427413702
  }
}