{ "modality": "text", "model": "sentence-transformers/all-MiniLM-L6-v2", "device": "cuda", "gpu": "NVIDIA GeForce RTX 5090", "embedding_dim": 384, "autobatch": { "min_batch": 8, "max_batch": 512, "start_batch": 32, "grow_free_fraction": 0.5, "shrink_free_fraction": 0.2, "grow_factor": 1.5, "shrink_factor": 0.5, "oom_factor": 0.5 }, "runs": [ { "num_workers": 1, "pre_run_counts": { "pending": 500, "in_progress": 0, "done": 0, "failed": 0, "requeued": 0 }, "total_processed": 500, "wall_seconds": 14.76, "throughput": 33.9, "note": "small-scale single-worker RTX 5090 run" } ] }