{ "eval_version": "sn-loop-eval-v1", "model": "sensenova-loop-nvfp4", "base_url": "http://127.0.0.1:30080/v1", "started": "2026-10-30T16:45:48-0700", "enable_thinking": false, "sampling": { "temperature": 0.0, "top_p": 1.0, "seed": 0, "max_tokens": { "ifeval": 1024, "gsm8k": 1024, "mmlu_pro": 1024, "humaneval": 1024, "tools": 512, "vqa": 64 } }, "concurrency": 16, "limit": null, "host": "tux", "data_sha256": { "ifeval": "3e9e9e7bd42ef54862f0c71c9a4e6f0b17a9c259f8f4b2b4e5fbcad917228118", "gsm8k": "5bb91d328c1ffc424d6679fe8edb36ed08152e4c75e75b343303e15cacc2da49", "mmlu_pro": "e9d8d5717def43a0b995710ab093a13de48b4cfb4db19cd7c66affacfb492ea7", "humaneval": "1e775e30085b890c19c48a583bfabb8f5f54598b5bf54796839b92e5bafdab9f", "tools": "037b2e288f4f063e8a2d994799b4a51e4d97740b1cf62da75844243749dbb08a", "vqa": "7d47303ecdf3386d04a852ab8b7e8be8d34137a1c4d10143331cc7175ce80d9d" }, "tasks": { "ifeval": { "n": 60, "correct": 45, "accuracy": 0.75, "wilson95": [ 0.6276792981955739, 0.8422347752873125 ], "n_errors": 0, "n_truncated": 4, "mean_completion_tokens": 272.6666666666667, "sum_completion_tokens": 16360, "instruction_level_acc": 0.7831325301204819, "n_instructions": 83, "wall_s": 196.0 }, "gsm8k": { "n": 80, "correct": 76, "accuracy": 0.95, "wilson95": [ 0.8783772371965909, 0.9803864456007073 ], "n_errors": 0, "n_truncated": 1, "mean_completion_tokens": 187.2, "sum_completion_tokens": 14976, "wall_s": 156.1 }, "mmlu_pro": { "n": 60, "correct": 41, "accuracy": 0.6833333333333333, "wilson95": [ 0.5576623287169538, 0.7869413251704964 ], "n_errors": 0, "n_truncated": 9, "mean_completion_tokens": 588.55, "sum_completion_tokens": 35313, "wall_s": 352.6 }, "humaneval": { "n": 40, "correct": 32, "accuracy": 0.8, "wilson95": [ 0.6524269352437855, 0.8950001032811363 ], "n_errors": 0, "n_truncated": 0, "mean_completion_tokens": 197.975, "sum_completion_tokens": 7919, "wall_s": 79.5 }, "tools": { "n": 20, "correct": 19, "accuracy": 0.95, "wilson95": [ 0.7638688045357189, 0.9911185512976646 ], "n_errors": 0, "n_truncated": 0, "mean_completion_tokens": 54.3, "sum_completion_tokens": 1086, "names_correct": 0.95, "parse_source_counts": { "server": 16, "none": 4 }, "format_well_formed": 1.0, "raw_tool_call_tags_in_content": 0, "by_category": { "irrelevance": "3/3", "multiple": "5/6", "parallel": "3/3", "simple": "8/8" }, "wall_s": 24.6 }, "vqa": { "n": 40, "correct": 36, "accuracy": 0.9, "wilson95": [ 0.7694822464406407, 0.9604204715925885 ], "n_errors": 0, "n_truncated": 0, "mean_completion_tokens": 4.075, "sum_completion_tokens": 163, "vqa_soft_acc": 0.8833333333333334, "wall_s": 25.5 } }, "server_models": { "object": "list", "data": [ { "id": "sensenova-loop-nvfp4", "object": "model", "created": 1793403948, "owned_by": "sglang", "root": "sensenova-loop-nvfp4", "parent": null, "max_model_len": 32768 } ] }, "wall_s": 834.3 }