GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
Silviase
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
Silviase
Model Tree
Qwen/Qwen3.5-2B
Input Modalities
Output Modalities
Supported Functionality
Full-image OCR uses CC-OCR character P/R/F1 and clipped order-free CER (image means), with NED as a diagnostic. Training set-F1 rewards are not full-image OCR evaluation. Paper metric version: jawildtext-metrics v2 (2026-09-28). Any earlier external exact/soft set-F1 report is superseded by this evaluation.
{ "dev": { "qa": { "n": 500, "mean_reward": 0.222 }, "evidence": { "n": 500, "mean_reward": 0.5189977128135022 }, "ocr": { "n": 500, "mean_reward": 0.8577615536872225 } }, "ocr": { "metrics_version": "jawildtext-metrics v2 (2026-09-28)", "images": 580, "capped": 59, "empty": 0, "precision": 0.8593098346746962, "recall": 0.8367336945230817, "f1": 0.8265248913630112, "ned": 0.3988669480288377, "cer": 0.21753777969872828, "character_micro_f1": 0.5787039885119365, "exact_region_recall": 0.40152395112205547, "cer_detail": { "clipped_images": 59, "exact_images": 475, "exact_rate": 0.8189655172413793, "mean_gap": 0.006714550350019816, "median_image_cer": 0.10166620575221239, "reference_characters": 263496, "pooled_cer": null, "substitutions": null, "deletions": null, "insertions": null } }, "qa": { "expected": 1025, "complete": 1025, "capped": 60, "metrics": { "content": { "value": 0.424390243902439, "ci95": [ 0.3926497590857178, 0.4560390483469203 ], "units": 1025, "images": 745, "excluded_units": 0, "valid_draws": 2000 }, "boxed": { "value": 0.4117073170731707, "ci95": [ 0.380809937595187, 0.4433648847884241 ], "units": 1025, "images": 745, "excluded_units": 0, "valid_draws": 2000 }, "format_valid": { "value": 0.9180487804878049, "ci95": [ 0.9015427373185994, 0.9341859583588255 ], "units": 1025, "images": 745, "excluded_units": 0, "valid_draws": 2000 } }, "scope": "All available outputs for this model; not a paired intersection", "repetitions": 2000, "seed": 42, "unit": "image cluster", "limitation": "Single training seed; intervals exclude training-seed variability" } }