{ "model_id": "gemini-2.5-flash", "provider_id": "gemini", "quality": { "accuracy": 85.5, "correct": 207, "total": 242, "model_id": "gemini-2.5-flash", "provider_id": "gemini", "sample_id": "conv-43" }, "retain": { "timestamp": "2026-02-20T13:54:07.284608+00:00", "model_id": "gemini-2.5-flash", "model_name": "gemini-2.5-flash", "provider_id": "gemini", "size_gb": 0.0, "dataset": "locomo_3k_50", "concurrency": 4, "wall_s": 759.4720120429993, "summary": { "success": 50, "total": 50, "wall_s": 759.472, "avg_latency_s": 58.878, "throughput_rps": 0.066, "completion_toks_s": 313.9, "total_toks_s": 609.2, "out_in_ratio": 1.063, "tokens_per_fact": 155.6 }, "tests": [ { "test_index": 1, "latency_s": 59.957480907440186, "num_facts": 43, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4444, "completion_tokens": 6200, "error": "" }, { "test_index": 2, "latency_s": 55.11748719215393, "num_facts": 40, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4516, "completion_tokens": 6160, "error": "" }, { "test_index": 3, "latency_s": 61.16433882713318, "num_facts": 34, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4353, "completion_tokens": 5227, "error": "" }, { "test_index": 4, "latency_s": 61.2218291759491, "num_facts": 38, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4625, "completion_tokens": 5629, "error": "" }, { "test_index": 5, "latency_s": 36.32994985580444, "num_facts": 13, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4467, "completion_tokens": 2293, "error": "" }, { "test_index": 6, "latency_s": 54.77700185775757, "num_facts": 24, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4537, "completion_tokens": 3727, "error": "" }, { "test_index": 7, "latency_s": 64.94908118247986, "num_facts": 44, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4444, "completion_tokens": 6937, "error": "" }, { "test_index": 8, "latency_s": 22.14191198348999, "num_facts": 13, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4621, "completion_tokens": 2051, "error": "" }, { "test_index": 9, "latency_s": 47.611276149749756, "num_facts": 20, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4300, "completion_tokens": 3682, "error": "" }, { "test_index": 10, "latency_s": 59.12394118309021, "num_facts": 33, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4298, "completion_tokens": 4862, "error": "" }, { "test_index": 11, "latency_s": 84.25143218040466, "num_facts": 59, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4464, "completion_tokens": 8022, "error": "" }, { "test_index": 12, "latency_s": 37.7795991897583, "num_facts": 32, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4605, "completion_tokens": 4284, "error": "" }, { "test_index": 13, "latency_s": 69.56385588645935, "num_facts": 29, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4378, "completion_tokens": 4859, "error": "" }, { "test_index": 14, "latency_s": 73.43493413925171, "num_facts": 45, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4556, "completion_tokens": 6219, "error": "" }, { "test_index": 15, "latency_s": 36.645387172698975, "num_facts": 21, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4590, "completion_tokens": 3263, "error": "" }, { "test_index": 16, "latency_s": 43.80622386932373, "num_facts": 21, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4501, "completion_tokens": 3644, "error": "" }, { "test_index": 17, "latency_s": 56.044952154159546, "num_facts": 30, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4605, "completion_tokens": 4529, "error": "" }, { "test_index": 18, "latency_s": 83.85160207748413, "num_facts": 24, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4570, "completion_tokens": 3535, "error": "" }, { "test_index": 19, "latency_s": 59.60909724235535, "num_facts": 33, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4360, "completion_tokens": 4864, "error": "" }, { "test_index": 20, "latency_s": 96.14667272567749, "num_facts": 33, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4495, "completion_tokens": 6106, "error": "" }, { "test_index": 21, "latency_s": 55.350512981414795, "num_facts": 27, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4605, "completion_tokens": 4253, "error": "" }, { "test_index": 22, "latency_s": 41.772764921188354, "num_facts": 24, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4501, "completion_tokens": 3733, "error": "" }, { "test_index": 23, "latency_s": 54.075897216796875, "num_facts": 22, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4516, "completion_tokens": 3596, "error": "" }, { "test_index": 24, "latency_s": 63.41429400444031, "num_facts": 43, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4537, "completion_tokens": 6209, "error": "" }, { "test_index": 25, "latency_s": 79.59361791610718, "num_facts": 46, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4495, "completion_tokens": 6954, "error": "" }, { "test_index": 26, "latency_s": 33.00387716293335, "num_facts": 17, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4570, "completion_tokens": 2819, "error": "" }, { "test_index": 27, "latency_s": 41.30866003036499, "num_facts": 29, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4298, "completion_tokens": 4651, "error": "" }, { "test_index": 28, "latency_s": 53.70343804359436, "num_facts": 25, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4501, "completion_tokens": 4428, "error": "" }, { "test_index": 29, "latency_s": 54.51125717163086, "num_facts": 16, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4625, "completion_tokens": 2810, "error": "" }, { "test_index": 30, "latency_s": 67.2601158618927, "num_facts": 45, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4409, "completion_tokens": 6423, "error": "" }, { "test_index": 31, "latency_s": 57.28098273277283, "num_facts": 15, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4467, "completion_tokens": 2611, "error": "" }, { "test_index": 32, "latency_s": 91.12612199783325, "num_facts": 31, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4516, "completion_tokens": 5151, "error": "" }, { "test_index": 33, "latency_s": 49.02624988555908, "num_facts": 33, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4598, "completion_tokens": 4493, "error": "" }, { "test_index": 34, "latency_s": 64.55250811576843, "num_facts": 41, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4464, "completion_tokens": 5389, "error": "" }, { "test_index": 35, "latency_s": 62.81263303756714, "num_facts": 18, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4590, "completion_tokens": 3163, "error": "" }, { "test_index": 36, "latency_s": 55.63943910598755, "num_facts": 30, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4556, "completion_tokens": 5026, "error": "" }, { "test_index": 37, "latency_s": 68.73234796524048, "num_facts": 46, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4409, "completion_tokens": 6790, "error": "" }, { "test_index": 38, "latency_s": 55.40490698814392, "num_facts": 21, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4300, "completion_tokens": 3866, "error": "" }, { "test_index": 39, "latency_s": 61.940956830978394, "num_facts": 28, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4378, "completion_tokens": 5034, "error": "" }, { "test_index": 40, "latency_s": 54.61986994743347, "num_facts": 30, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4360, "completion_tokens": 4821, "error": "" }, { "test_index": 41, "latency_s": 60.04973125457764, "num_facts": 22, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4378, "completion_tokens": 4237, "error": "" }, { "test_index": 42, "latency_s": 49.98315906524658, "num_facts": 31, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4298, "completion_tokens": 5026, "error": "" }, { "test_index": 43, "latency_s": 53.50798320770264, "num_facts": 38, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4625, "completion_tokens": 6274, "error": "" }, { "test_index": 44, "latency_s": 71.83177995681763, "num_facts": 38, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4621, "completion_tokens": 5565, "error": "" }, { "test_index": 45, "latency_s": 64.41385817527771, "num_facts": 41, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4409, "completion_tokens": 6052, "error": "" }, { "test_index": 46, "latency_s": 35.615665912628174, "num_facts": 17, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4467, "completion_tokens": 2455, "error": "" }, { "test_index": 47, "latency_s": 81.52653908729553, "num_facts": 46, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4621, "completion_tokens": 6091, "error": "" }, { "test_index": 48, "latency_s": 42.70303726196289, "num_facts": 17, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4598, "completion_tokens": 2886, "error": "" }, { "test_index": 49, "latency_s": 92.09746098518372, "num_facts": 35, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4495, "completion_tokens": 6409, "error": "" }, { "test_index": 50, "latency_s": 63.49572706222534, "num_facts": 31, "valid_json": true, "success": true, "retries": 0, "prompt_tokens": 4353, "completion_tokens": 5085, "error": "" } ] }, "reflect": { "accuracy": 85.5, "correct": 207, "total": 242, "avg_latency_s": 4.447, "model_id": "gemini-2.5-flash", "provider_id": "gemini", "sample_id": "conv-43" } }