{ "purpose": "selected local Core ML extraction parity and latency, not a Decision Index score", "source_model": "fastino/gliner2.5-multi-v1", "source_revision": "a221b77a8baf4a613b8f8652661d41fa10a5641e", "runtime": { "coremltools": "9.0", "gliner2": "2.0.0", "mac": "M5 Pro, 24 GB, macOS 27.0" }, "bucket": "L128/W64/Q8/K8; candidate C192, explicit S64, relation R4/P256, records F8/C192/I1536", "fixtures": { "fp32": { "matched_structures": 15, "fixture_count": 15, "failed_cases": [], "maximum_confidence_error": 4.887580871582031e-06 }, "fp16": { "matched_structures": 14, "fixture_count": 15, "failed_cases": [ "record_latent" ], "maximum_confidence_error": 0.037593305110931396 } }, "latency": { "fp32_all": { "p50_ms": 10.563604475464672, "p95_ms": 12.338125030510128, "warmup": 20, "iterations": 200, "shape": "L128/W64/Q8/C192" }, "fp32_cpu_and_gpu": { "p50_ms": 10.022375005064532, "p95_ms": 10.64095797482878, "warmup": 20, "iterations": 200, "shape": "L128/W64/Q8/C192" }, "fp32_cpu_and_neural_engine": { "p50_ms": 36.07562501565553, "p95_ms": 52.11695801699534, "warmup": 20, "iterations": 200, "shape": "L128/W64/Q8/C192" }, "fp32_cpu_only": { "p50_ms": 36.11889548483305, "p95_ms": 62.56379099795595, "warmup": 20, "iterations": 200, "shape": "L128/W64/Q8/C192" } }, "feature_compute_plan": { "fp32-all": { "cpu_percent": 0.0, "gpu_percent": 100.0, "ane_percent": 0.0 }, "fp32-cpu_and_neural_engine": { "cpu_percent": 100.0, "gpu_percent": 0.0, "ane_percent": 0.0 } }, "compression": { "attempt": "LUT8 per-tensor k-means on FP32 feature package using scikit-learn 1.5.1", "result": "stopped after two minutes with zero graph operations processed and repeated k-means divide-by-zero/overflow warnings", "release": false }, "published_extraction_precisions": [ "fp32" ], "limitations": [ "fixed shape; requests over capacity fail", "small selected real-text fixtures only; no full Decision Index score", "current pinned source revision is not established as the historical evaluation checkpoint" ] }