Upload README.md with huggingface_hub

README.md

@@ -42,3 +42,35 @@ Fine-tuned `microsoft/deberta-v3-small` — 3-class classifier predicting which
   being a minority class
 - Best checkpoint selected by validation F1 macro (monotonically improved
   every epoch: 0.708 → 0.788 → 0.792 → 0.798)
+
+## Comprehensive Benchmark
+
+### Per-Source Accuracy (test set)
+
+| Source | n | Accuracy |
+|---|---|---|
+| flan | 4,108 | 97.4% |
+| openhermes | 5,929 | 87.5% |
+| evol | 2,479 | 85.0% |
+| sharegpt | 1,431 | 78.3% |
+
+### Calibration
+
+Expected Calibration Error (ECE): **0.053**.
+
+### Inference Latency (single-sample, batch=1)
+
+| Device | p50 | p95 | p99 |
+|---|---|---|---|
+| GPU (T4) | 15.8ms | 23.6ms | 44.0ms |
+| CPU | 158.2ms | 412.0ms | 483.4ms |
+
+GPU inference is ~12x faster than CPU.
+
+### Error Analysis
+
+- Overall accuracy: 89.0% (test set, n=13,947)
+- Critical failure rate (frontier prompt routed to `fast`): **3.6%**
+- Over-provisioning rate (fast prompt routed to `frontier`): **0.17%**
+
+The error asymmetry favors safety: the model is far more likely to over-provision an easy prompt than under-provision a hard one.