diff --git a/.modelhub_state/architecture_compatibility_blacklist.json b/.modelhub_state/architecture_compatibility_blacklist.json index 771dd290a1..ed5abf7624 100644 --- a/.modelhub_state/architecture_compatibility_blacklist.json +++ b/.modelhub_state/architecture_compatibility_blacklist.json @@ -2871,7 +2871,7 @@ "taskType": "text-generation" } }, - "generatedAt": "2026-09-21T23:59:03.775821+00:00", + "generatedAt": "2026-09-22T00:00:30.378708+00:00", "summary": { "activeBlockCount": 144, "byGpuFramework": { diff --git a/.modelhub_state/market_intelligence.json b/.modelhub_state/market_intelligence.json index c451d42312..7098a4cf2d 100644 --- a/.modelhub_state/market_intelligence.json +++ b/.modelhub_state/market_intelligence.json @@ -434,121 +434,121 @@ } }, "frameworkUpdatedAt": null, - "generatedAt": "2026-09-21T23:58:34.577341+00:00", + "generatedAt": "2026-09-22T00:00:41.397746+00:00", "gpuStats": { "Ascend_910-b3": { "available": true, - "backlogHours": 979.1666666666666, + "backlogHours": 972.2068965517241, "canVerify": true, "error": null, "gpu": "Ascend_910-b3", "healthFactor": 1.0, "maxConcurrentTasks": 8, - "qualityFactor": 0.5892242841853491, - "queueFactor": 0.9972810192940773, - "queueWeight": 0.9972810192940773, - "recentSuccess": 38, - "recentSuccessRate": 0.2638888888888889, - "recentTerminal": 144, - "recentWilsonLowerBound": 0.1987136611938409, + "qualityFactor": 0.6523352278842933, + "queueFactor": 0.9978334396537183, + "queueWeight": 0.9978334396537183, + "recentSuccess": 39, + "recentSuccessRate": 0.2689655172413793, + "recentTerminal": 145, + "recentWilsonLowerBound": 0.2034414774869696, "running": 8, - "selectionWeight": 0.5876221947251881, + "selectionWeight": 0.6509219042470765, "stale": false, "submissionEligible": true, - "throughputPerHour": 24.0, - "waiting": 23500 + "throughputPerHour": 24.166666666666668, + "waiting": 23495 }, "Ascend_910-b4": { "available": true, - "backlogHours": 1220.8489208633093, + "backlogHours": 1238.7591240875913, "canVerify": true, "error": null, "gpu": "Ascend_910-b4", "healthFactor": 1.0, "maxConcurrentTasks": 8, - "qualityFactor": 1.7392347622955056, - "queueFactor": 0.9648210229809128, - "queueWeight": 0.9648210229809128, - "recentSuccess": 56, - "recentSuccessRate": 0.4028776978417266, - "recentTerminal": 139, - "recentWilsonLowerBound": 0.32501196368104285, + "qualityFactor": 1.7222310281631115, + "queueFactor": 0.9622187794967229, + "queueWeight": 0.9622187794967229, + "recentSuccess": 54, + "recentSuccessRate": 0.39416058394160586, + "recentTerminal": 137, + "recentWilsonLowerBound": 0.3162898229154807, "running": 8, - "selectionWeight": 1.6780502625619143, + "selectionWeight": 1.6571630379304951, "stale": false, "submissionEligible": true, - "throughputPerHour": 23.166666666666668, - "waiting": 28283 + "throughputPerHour": 22.833333333333332, + "waiting": 28285 }, "Biren_166m": { "available": true, - "backlogHours": 180.53571428571428, + "backlogHours": 176.19767441860463, "canVerify": true, "error": null, "gpu": "Biren_166m", "healthFactor": 1.0, "maxConcurrentTasks": 8, - "qualityFactor": 0.3352575809456031, - "queueFactor": 1.2851729534667333, - "queueWeight": 1.2851729534667333, - "recentSuccess": 35, - "recentSuccessRate": 0.20833333333333334, - "recentTerminal": 168, - "recentWilsonLowerBound": 0.15378310320374775, + "qualityFactor": 0.38608615001372504, + "queueFactor": 1.2892045596848076, + "queueWeight": 1.2892045596848076, + "recentSuccess": 37, + "recentSuccessRate": 0.21511627906976744, + "recentTerminal": 172, + "recentWilsonLowerBound": 0.16028767638500904, "running": 8, - "selectionWeight": 0.43086397547597316, + "selectionWeight": 0.49774402502884696, "stale": false, "submissionEligible": true, - "throughputPerHour": 28.0, - "waiting": 5055 + "throughputPerHour": 28.666666666666668, + "waiting": 5051 }, "Cambricon_mlu-370-x4": { "available": true, - "backlogHours": 2101.6744186046512, + "backlogHours": 2151.714285714286, "canVerify": true, "error": null, "gpu": "Cambricon_mlu-370-x4", "healthFactor": 1.0, "maxConcurrentTasks": 7, - "qualityFactor": 0.6305466178638047, - "queueFactor": 0.8893264943907511, - "queueWeight": 0.8893264943907511, + "qualityFactor": 0.7003980992558062, + "queueFactor": 0.8857357381347704, + "queueWeight": 0.8857357381347704, "recentSuccess": 14, - "recentSuccessRate": 0.32558139534883723, - "recentTerminal": 43, - "recentWilsonLowerBound": 0.20493115181779817, - "running": 8, - "selectionWeight": 0.560761813214762, + "recentSuccessRate": 0.3333333333333333, + "recentTerminal": 42, + "recentWilsonLowerBound": 0.21012280761972318, + "running": 7, + "selectionWeight": 0.6203676274325317, "stale": false, "submissionEligible": true, - "throughputPerHour": 7.166666666666667, + "throughputPerHour": 7.0, "waiting": 15062 }, "Cambricon_mlu-370-x8": { "available": true, - "backlogHours": 483.1578947368421, + "backlogHours": 462.95798319327736, "canVerify": true, "error": null, "gpu": "Cambricon_mlu-370-x8", "healthFactor": 1.0, "maxConcurrentTasks": 7, - "qualityFactor": 0.2657477529253461, - "queueFactor": 1.10874757883722, - "queueWeight": 1.10874757883722, - "recentSuccess": 23, - "recentSuccessRate": 0.20175438596491227, - "recentTerminal": 114, - "recentWilsonLowerBound": 0.13836966525548045, + "qualityFactor": 0.2841595722788314, + "queueFactor": 1.1152972260971574, + "queueWeight": 1.1152972260971574, + "recentSuccess": 24, + "recentSuccessRate": 0.20168067226890757, + "recentTerminal": 119, + "recentWilsonLowerBound": 0.13944089576973434, "running": 7, - "selectionWeight": 0.2946471776374092, + "selectionWeight": 0.31692238273153533, "stale": false, "submissionEligible": true, - "throughputPerHour": 19.0, - "waiting": 9180 + "throughputPerHour": 19.833333333333332, + "waiting": 9182 }, "Iluvatar_bi-100": { "available": true, - "backlogHours": 40.56410256410256, + "backlogHours": 40.02253521126761, "canVerify": true, "error": null, "gpu": "Iluvatar_bi-100", @@ -558,125 +558,125 @@ "queueFactor": 1.3, "queueWeight": 1.3, "recentSuccess": 10, - "recentSuccessRate": 0.02849002849002849, - "recentTerminal": 351, - "recentWilsonLowerBound": 0.015547266406963859, - "running": 25, + "recentSuccessRate": 0.028169014084507043, + "recentTerminal": 355, + "recentWilsonLowerBound": 0.015371278342861802, + "running": 24, "selectionWeight": 0.065, "stale": false, "submissionEligible": false, - "throughputPerHour": 58.5, - "waiting": 2373 + "throughputPerHour": 59.166666666666664, + "waiting": 2368 }, "Iluvatar_bi-150": { "available": true, - "backlogHours": 117.44578313253012, + "backlogHours": 113.34883720930233, "canVerify": true, "error": null, "gpu": "Iluvatar_bi-150", "healthFactor": 1.0, "maxConcurrentTasks": 50, - "qualityFactor": 0.05149534516257742, + "qualityFactor": 0.05001698855467185, "queueFactor": 1.3, "queueWeight": 1.3, "recentSuccess": 24, - "recentSuccessRate": 0.0963855421686747, - "recentTerminal": 249, - "recentWilsonLowerBound": 0.06562746137893503, - "running": 3, - "selectionWeight": 0.06694394871135065, + "recentSuccessRate": 0.09302325581395349, + "recentTerminal": 258, + "recentWilsonLowerBound": 0.06330831170978082, + "running": 4, + "selectionWeight": 0.06502208512107341, "stale": false, "submissionEligible": true, - "throughputPerHour": 41.5, + "throughputPerHour": 43.0, "waiting": 4874 }, "Iluvatar_mrv-100": { "available": true, - "backlogHours": 2830.235294117647, + "backlogHours": 2749.2000000000003, "canVerify": true, "error": null, "gpu": "Iluvatar_mrv-100", "healthFactor": 1.0, "maxConcurrentTasks": 2, - "qualityFactor": 0.5420321664225047, - "queueFactor": 0.8504967632319744, - "queueWeight": 0.8504967632319744, - "recentSuccess": 11, - "recentSuccessRate": 0.3235294117647059, - "recentTerminal": 34, - "recentWilsonLowerBound": 0.19131451326458573, + "qualityFactor": 0.6872091096013734, + "queueFactor": 0.8537700353810594, + "queueWeight": 0.8537700353810594, + "recentSuccess": 12, + "recentSuccessRate": 0.34285714285714286, + "recentTerminal": 35, + "recentWilsonLowerBound": 0.20831495282688925, "running": 2, - "selectionWeight": 0.46099660310995516, + "selectionWeight": 0.5867185458185509, "stale": false, "submissionEligible": true, - "throughputPerHour": 5.666666666666667, - "waiting": 16038 + "throughputPerHour": 5.833333333333333, + "waiting": 16037 }, "Kunlunxin_p-800": { "available": true, - "backlogHours": 658.75, + "backlogHours": 645.3061224489796, "canVerify": true, "error": null, "gpu": "Kunlunxin_p-800", "healthFactor": 1.0, "maxConcurrentTasks": 8, - "qualityFactor": 0.6665162636305064, - "queueFactor": 1.0583710508514597, - "queueWeight": 1.0583710508514597, - "recentSuccess": 28, - "recentSuccessRate": 0.2916666666666667, - "recentTerminal": 96, - "recentWilsonLowerBound": 0.21016461419964405, + "qualityFactor": 0.7338800191496985, + "queueFactor": 1.061101652370322, + "queueWeight": 1.061101652370322, + "recentSuccess": 29, + "recentSuccessRate": 0.29591836734693877, + "recentTerminal": 98, + "recentWilsonLowerBound": 0.21463049371609197, "running": 8, - "selectionWeight": 0.7054215183482077, + "selectionWeight": 0.7787213009613086, "stale": false, "submissionEligible": true, - "throughputPerHour": 16.0, + "throughputPerHour": 16.333333333333332, "waiting": 10540 }, "MetaX_c-500": { "available": true, - "backlogHours": 669.8969072164948, + "backlogHours": 669.7113402061855, "canVerify": true, "error": null, "gpu": "MetaX_c-500", "healthFactor": 1.0, "maxConcurrentTasks": 4, - "qualityFactor": 0.7823269110878763, - "queueFactor": 1.0557105232681236, - "queueWeight": 1.0557105232681236, - "recentSuccess": 30, - "recentSuccessRate": 0.30927835051546393, + "qualityFactor": 0.7514035986459311, + "queueFactor": 1.0552095443184655, + "queueWeight": 1.0552095443184655, + "recentSuccess": 29, + "recentSuccessRate": 0.29896907216494845, "recentTerminal": 97, - "recentWilsonLowerBound": 0.22604022957947018, + "recentWilsonLowerBound": 0.21694502760133394, "running": 4, - "selectionWeight": 0.8259107526713168, + "selectionWeight": 0.7928882489264281, "stale": false, "submissionEligible": true, "throughputPerHour": 16.166666666666668, - "waiting": 10830 + "waiting": 10827 }, "Mthreads_s4000": { "available": true, - "backlogHours": 2740.375, + "backlogHours": 2684.8163265306125, "canVerify": true, "error": null, "gpu": "Mthreads_s4000", "healthFactor": 1.0, "maxConcurrentTasks": 4, "qualityFactor": 2.5, - "queueFactor": 0.85462294178185, - "queueWeight": 0.85462294178185, + "queueFactor": 0.8568102861718285, + "queueWeight": 0.8568102861718285, "recentSuccess": 25, - "recentSuccessRate": 0.5208333333333334, - "recentTerminal": 48, - "recentWilsonLowerBound": 0.38328995635984375, + "recentSuccessRate": 0.5102040816326531, + "recentTerminal": 49, + "recentWilsonLowerBound": 0.3746733153340922, "running": 4, - "selectionWeight": 2.136557354454625, + "selectionWeight": 2.1420257154295714, "stale": false, "submissionEligible": true, - "throughputPerHour": 8.0, - "waiting": 21923 + "throughputPerHour": 8.166666666666666, + "waiting": 21926 }, "Sunrise_pt-200-x1": { "available": true, @@ -686,15 +686,15 @@ "gpu": "Sunrise_pt-200-x1", "healthFactor": 1.0, "maxConcurrentTasks": 2, - "qualityFactor": 0.38417183556728784, - "queueFactor": 0.7985547408121914, - "queueWeight": 0.7985547408121914, + "qualityFactor": 0.4038776400296829, + "queueFactor": 0.798142623969382, + "queueWeight": 0.798142623969382, "recentSuccess": 7, "recentSuccessRate": 0.3181818181818182, "recentTerminal": 22, "recentWilsonLowerBound": 0.16360387354692477, "running": 2, - "selectionWeight": 0.30678224057877934, + "selectionWeight": 0.3223519593758526, "stale": false, "submissionEligible": true, "throughputPerHour": 3.6666666666666665, @@ -709,14 +709,14 @@ "healthFactor": 1.0, "maxConcurrentTasks": 16, "qualityFactor": 0.05, - "queueFactor": 0.9176860541809757, - "queueWeight": 0.9176860541809757, + "queueFactor": 0.917212456242086, + "queueWeight": 0.917212456242086, "recentSuccess": 6, "recentSuccessRate": 0.11764705882352941, "recentTerminal": 51, "recentWilsonLowerBound": 0.05504998764172767, - "running": 17, - "selectionWeight": 0.04588430270904879, + "running": 16, + "selectionWeight": 0.045860622812104304, "stale": false, "submissionEligible": true, "throughputPerHour": 8.5, @@ -724,30 +724,30 @@ }, "hygon_k100-ai": { "available": true, - "backlogHours": 943.9411764705883, + "backlogHours": 944.2941176470588, "canVerify": true, "error": null, "gpu": "hygon_k100-ai", "healthFactor": 1.0, "maxConcurrentTasks": 6, - "qualityFactor": 0.20371044052485024, - "queueFactor": 1.0027768673215995, - "queueWeight": 1.0027768673215995, + "qualityFactor": 0.214159613879798, + "queueFactor": 1.0022031559689415, + "queueWeight": 1.0022031559689415, "recentSuccess": 19, "recentSuccessRate": 0.18627450980392157, "recentTerminal": 102, "recentWilsonLowerBound": 0.12261984082072329, - "running": 7, - "selectionWeight": 0.20427611739021234, + "running": 6, + "selectionWeight": 0.2146314409114235, "stale": false, "submissionEligible": true, "throughputPerHour": 17.0, - "waiting": 16047 + "waiting": 16053 } }, - "queueAttemptedAt": "2026-09-21T23:49:41.847306+00:00", + "queueAttemptedAt": "2026-09-22T00:00:41.397746+00:00", "queueError": null, - "queueUpdatedAt": "2026-09-21T23:49:41.847306+00:00", + "queueUpdatedAt": "2026-09-22T00:00:41.397746+00:00", "supportedGpus": [ "Vastai_va16", "Kunlunxin_p-800", diff --git a/.modelhub_state/official_capabilities.json b/.modelhub_state/official_capabilities.json index 6b304cd967..a37cf306b0 100644 --- a/.modelhub_state/official_capabilities.json +++ b/.modelhub_state/official_capabilities.json @@ -1,5 +1,5 @@ { - "catalogUpdatedAt": "2026-09-21T23:58:34.577341+00:00", + "catalogUpdatedAt": "2026-09-22T00:00:41.397746+00:00", "configuredTaskTypes": [ "text-generation" ], @@ -56,7 +56,7 @@ "time-series-forecasting" ], "errors": [], - "generatedAt": "2026-09-21T23:58:43.357238+00:00", + "generatedAt": "2026-09-22T00:00:41.397746+00:00", "gpuCatalog": { "Ascend_910-b3": { "canVerify": true, @@ -6813,6 +6813,6 @@ "updateTime": "2025-12-22 08:59:53" } ], - "taskTreeUpdatedAt": "2026-09-21T23:58:34.577341+00:00", + "taskTreeUpdatedAt": "2026-09-22T00:00:41.397746+00:00", "version": 1 } diff --git a/.modelhub_state/outcome_checkpoint.json b/.modelhub_state/outcome_checkpoint.json index a90e13a789..b2d7199c6c 100644 --- a/.modelhub_state/outcome_checkpoint.json +++ b/.modelhub_state/outcome_checkpoint.json @@ -1,6 +1,6 @@ { - "generatedAt": "2026-09-21T23:59:03.698072+00:00", - "lastSyncTime": "2026-09-21T23:59:03.362604+00:00", + "generatedAt": "2026-09-22T00:00:30.301244+00:00", + "lastSyncTime": "2026-09-22T00:00:30.059018+00:00", "recentLimit": 300, "report": { "architectureCompatibilityBlocks": { @@ -4529,24 +4529,24 @@ "decisionSuccessRate": 0.1, "decisionTotal": 20, "failureBreakdown": { - "ambiguous_runtime": 2, + "ambiguous_runtime": 3, "backend_operator": 2, "framework_architecture_unsupported": 14, "repository_structure": 2, "参数/模板问题": 14 }, - "failureCount": 34, - "failureRate": 0.9444, + "failureCount": 35, + "failureRate": 0.9459, "framework": "vllm_tokenizer_patch", "pendingCount": 0, "pendingRate": 0.0, "platformFailureCount": 0, "successCount": 2, - "successRate": 0.0556, + "successRate": 0.0541, "targetGpu": "Kunlunxin_p-800", "taskType": "text-generation", - "total": 36, - "unresolvedFailureCount": 16 + "total": 37, + "unresolvedFailureCount": 17 }, "Kunlunxin_p-800|vllm|text-generation": { "attributableFailureCount": 1, @@ -5678,7 +5678,7 @@ "decisionSuccessRate": 0.0482, "decisionTotal": 83, "failureBreakdown": { - "ambiguous_runtime": 94, + "ambiguous_runtime": 95, "backend_operator": 2, "context_length": 3, "framework_architecture_unsupported": 66, @@ -5689,18 +5689,18 @@ "tokenizer_compatibility": 2, "参数/模板问题": 28 }, - "failureCount": 202, - "failureRate": 0.9806, + "failureCount": 203, + "failureRate": 0.9807, "pendingCount": 0, "pendingRate": 0.0, "platformFailureCount": 1, "successCount": 4, - "successRate": 0.0194, - "total": 206, - "unresolvedFailureCount": 122 + "successRate": 0.0193, + "total": 207, + "unresolvedFailureCount": 123 } }, - "generatedAt": "2026-09-21T23:59:03.686238+00:00", + "generatedAt": "2026-09-22T00:00:30.289957+00:00", "gpuSummaries": { "Ascend_910-b3": { "attributableFailureCount": 99, @@ -5930,7 +5930,7 @@ "decisionSuccessRate": 0.0909, "decisionTotal": 22, "failureBreakdown": { - "ambiguous_runtime": 189, + "ambiguous_runtime": 190, "backend_operator": 2, "framework_architecture_unsupported": 15, "memory_capacity": 1, @@ -5939,15 +5939,15 @@ "日志缺失": 1, "验证失败": 23 }, - "failureCount": 254, + "failureCount": 255, "failureRate": 0.9922, "pendingCount": 0, "pendingRate": 0.0, "platformFailureCount": 0, "successCount": 2, "successRate": 0.0078, - "total": 256, - "unresolvedFailureCount": 234 + "total": 257, + "unresolvedFailureCount": 235 }, "Kunlunxin_r-200-8f": { "attributableFailureCount": 0, @@ -15023,9 +15023,11 @@ "decisionFailureRate": 0.0, "decisionSuccessRate": 1.0, "decisionTotal": 2, - "failureBreakdown": {}, - "failureCount": 0, - "failureRate": 0.0, + "failureBreakdown": { + "ambiguous_runtime": 1 + }, + "failureCount": 1, + "failureRate": 0.3333, "framework": "vllm_tokenizer_patch", "modelType": "llama", "pendingCount": 0, @@ -15033,11 +15035,11 @@ "platformFailureCount": 0, "quantizationMethod": "awq", "successCount": 2, - "successRate": 1.0, + "successRate": 0.6667, "targetGpu": "Kunlunxin_p-800", "taskType": "text-generation", - "total": 2, - "unresolvedFailureCount": 0 + "total": 3, + "unresolvedFailureCount": 1 }, "Kunlunxin_p-800|vllm_tokenizer_patch|text-generation|llama|compressed-tensors": { "attributableFailureCount": 1, @@ -34750,9 +34752,11 @@ "decisionFailureRate": 0.0, "decisionSuccessRate": 1.0, "decisionTotal": 1, - "failureBreakdown": {}, - "failureCount": 0, - "failureRate": 0.0, + "failureBreakdown": { + "ambiguous_runtime": 1 + }, + "failureCount": 1, + "failureRate": 0.5, "framework": "vllm_tokenizer_patch", "loadSizeLog2Bucket": 32, "modelType": "llama", @@ -34761,11 +34765,11 @@ "platformFailureCount": 0, "quantizationMethod": "awq", "successCount": 1, - "successRate": 1.0, + "successRate": 0.5, "targetGpu": "Kunlunxin_p-800", "taskType": "text-generation", - "total": 1, - "unresolvedFailureCount": 0 + "total": 2, + "unresolvedFailureCount": 1 }, "Kunlunxin_p-800|vllm_tokenizer_patch|text-generation|llama|compressed-tensors|35": { "attributableFailureCount": 1, @@ -38789,15 +38793,15 @@ "unresolvedFailureCount": 0 } }, - "terminalRecords": 16446, - "totalRecords": 16640, + "terminalRecords": 16447, + "totalRecords": 16641, "totals": { "attributableFailureCount": 5925, "decisionFailureRate": 0.8619, "decisionSuccessRate": 0.1381, "decisionTotal": 6874, "failureBreakdown": { - "ambiguous_runtime": 4059, + "ambiguous_runtime": 4060, "architecture_compatibility": 212, "attention_backend": 1, "backend_operator": 105, @@ -38813,31 +38817,31 @@ "日志缺失": 719, "验证失败": 674 }, - "failureCount": 15497, + "failureCount": 15498, "failureRate": 0.9423, "pendingCount": 0, "pendingRate": 0.0, "platformFailureCount": 925, "successCount": 949, "successRate": 0.0577, - "total": 16446, - "unresolvedFailureCount": 8647 + "total": 16447, + "unresolvedFailureCount": 8648 }, "warnings": [ "GPU Iluvatar_mrv-100 本地统计失败率偏高（≥50%），建议重点关注。", "GPU Mthreads_s4000 本地统计失败率偏高（≥50%），建议重点关注。", - "GPU Cambricon_mlu-370-x8 本地统计失败率偏高（≥50%），建议重点关注。", "GPU Kunlunxin_p-800 本地统计失败率偏高（≥50%），建议重点关注。", - "GPU MetaX_c-500 本地统计失败率偏高（≥50%），建议重点关注。", "GPU Biren_166m 本地统计失败率偏高（≥50%），建议重点关注。", "GPU Iluvatar_bi-100 本地统计失败率偏高（≥50%），建议重点关注。", "GPU Sunrise_pt-200-x1 本地统计失败率偏高（≥50%），建议重点关注。", "GPU Ascend_910-b3 本地统计失败率偏高（≥50%），建议重点关注。", + "GPU Vastai_va16 本地统计失败率偏高（≥50%），建议重点关注。", "GPU Cambricon_mlu-370-x4 本地统计失败率偏高（≥50%），建议重点关注。", "GPU Iluvatar_bi-150 本地统计失败率偏高（≥50%），建议重点关注。", + "GPU Cambricon_mlu-370-x8 本地统计失败率偏高（≥50%），建议重点关注。", "GPU hygon_k100-ai 本地统计失败率偏高（≥50%），建议重点关注。", "GPU Ascend_910-b4 本地统计失败率偏高（≥50%），建议重点关注。", - "GPU Vastai_va16 本地统计失败率偏高（≥50%），建议重点关注。", + "GPU MetaX_c-500 本地统计失败率偏高（≥50%），建议重点关注。", "组合 Iluvatar_bi-150|vllm|visual-multi-modal 近期失败集中，建议降低该 GPU+框架的提交优先级。", "组合 Sunrise_pt-200-x1|vllm_fix_tokenizer|text-generation 近期失败集中，建议降低该 GPU+框架的提交优先级。", "组合 Cambricon_mlu-370-x8|unknown|text-generation 近期失败集中，建议降低该 GPU+框架的提交优先级。", @@ -38877,6 +38881,6 @@ ] }, "storageMode": "decision_state_only", - "summarizedRecords": 16640, + "summarizedRecords": 16641, "version": 1 } diff --git a/.modelhub_state/recovery_intents.jsonl b/.modelhub_state/recovery_intents.jsonl index c6489fd6d9..0d499edce6 100644 --- a/.modelhub_state/recovery_intents.jsonl +++ b/.modelhub_state/recovery_intents.jsonl @@ -2343,6 +2343,7 @@ {"batchId": "8e56ee5e993148539afedf340419216a", "completedAt": "2026-09-21T23:50:56.286873+00:00", "configFingerprint": "533968aac3b9c90272dfbe2866791f782e077f0dceee5b6ffffce48ba6674a10", "configSource": "modelhub_live", "createdAt": "2026-09-21T23:50:55.193948+00:00", "framework": "vllm", "intentId": "c81bca32e7b04b129042e7b5d0fcabab", "lastModified": "2026-09-21T18:42:39+00:00", "modelAddress": "https://modelscope.cn/models/nkkbr/Mini-K3-1H-mamba2-n128-g8-v1_B", "reason": null, "reconciledAt": "2026-09-21T23:59:18.464084+00:00", "repoId": "nkkbr/Mini-K3-1H-mamba2-n128-g8-v1_B", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "recovered_active", "targetGpu": "MetaX_c-500", "taskId": "5009165", "taskType": "text-generation"} {"batchId": "42d9d4c9dbe642b2b8bcb041a4502552", "completedAt": "2026-09-21T23:54:44.173934+00:00", "configFingerprint": "71d6e16091ba0289f01acffc8e39bb8c25703488dc49b9e390e08951fe1f2b33", "configSource": "modelhub_live", "createdAt": "2026-09-21T23:54:42.085677+00:00", "framework": "vllm", "intentId": "1ed5b82e591e4822b000359f429c22a2", "lastModified": "2026-09-21T18:43:38+00:00", "modelAddress": "https://modelscope.cn/models/nkkbr/Mini-K3-1H-decay-g2-v2_C", "reason": null, "reconciledAt": "2026-09-21T23:59:18.464802+00:00", "repoId": "nkkbr/Mini-K3-1H-decay-g2-v2_C", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "recovered_active", "targetGpu": "Iluvatar_mrv-100", "taskId": "5009242", "taskType": "text-generation"} {"batchId": "c3f3b8ca5a434fea8a529f52f3307342", "completedAt": "2026-09-21T23:58:44.522596+00:00", "configFingerprint": "2e9cebc01b20081f0839fa18fe9ca4f7d059509f235699027201fe328177b304", "configSource": "modelhub_live", "createdAt": "2026-09-21T23:58:43.496878+00:00", "framework": "vllm_fix_tokenizer", "intentId": "3615896687d546eaa9e73087955579f5", "lastModified": "2026-09-21T18:44:33+00:00", "modelAddress": "https://modelscope.cn/models/nkkbr/Mini-K3-1H-attnres-block6-v1_C", "reason": null, "reconciledAt": "2026-09-21T23:59:18.465265+00:00", "repoId": "nkkbr/Mini-K3-1H-attnres-block6-v1_C", "safeConfigVector": {"dtype": "-", "gpuNum": 1, "maxModelLen": 4096}, "status": "recovered_active", "targetGpu": "Sunrise_pt-200-x1", "taskId": "5009282", "taskType": "text-generation"} +{"batchId": "77c60f39e77d4f3ea4c9f43602976ae4", "configFingerprint": "2e9cebc01b20081f0839fa18fe9ca4f7d059509f235699027201fe328177b304", "configSource": "modelhub_live", "createdAt": "2026-09-22T00:00:42.698997+00:00", "framework": "vllm_fix_tokenizer", "intentId": "f1e9ea41f3e74c27ac03cb7e5de8cdb5", "lastModified": "2026-09-21T18:45:44+00:00", "modelAddress": "https://modelscope.cn/models/nkkbr/Mini-K3-1H-decay-g16-v2_B", "repoId": "nkkbr/Mini-K3-1H-decay-g16-v2_B", "safeConfigVector": {"dtype": "-", "gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Sunrise_pt-200-x1", "taskType": "text-generation"} {"batchId": "24082fb4e4bc48b988e1e4deac30068b", "completedAt": "2026-09-21T20:36:17.828196+00:00", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-21T20:36:00.461127+00:00", "framework": "vllm_fix_tokenizer", "intentId": "7ef98d8672514b6cb8bcf83850090ffe", "repoId": "nkkbr/Mini-K3-1H-attnres-block6-v1_C", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "capacity_deferred", "targetGpu": "Biren_166m", "taskId": null, "taskType": "text-generation"} {"batchId": "f8d3f1991720478ea905a5ecf7d43601", "completedAt": "2026-09-21T19:05:00.691781+00:00", "configFingerprint": "2ebd1d3ee8ed952dd08e24120ad5d7342c9a018f55c478225a242b664721e8d0", "configSource": "modelhub_live", "createdAt": "2026-09-21T19:00:55.137064+00:00", "framework": "vllm", "intentId": "64b834dd640742a0b32535bde02b94ad", "repoId": "CohereLabs/tiny-aya-base-32K", "safeConfigVector": {"dtype": "-", "gpuNum": 1, "maxModelLen": 2048}, "status": "age_policy_deferred", "targetGpu": "Ascend_910-b4", "taskId": null, "taskType": "text-generation"} {"batchId": "f8d3f1991720478ea905a5ecf7d43601", "completedAt": "2026-09-21T19:05:00.691779+00:00", "configFingerprint": "2ebd1d3ee8ed952dd08e24120ad5d7342c9a018f55c478225a242b664721e8d0", "configSource": "modelhub_live", "createdAt": "2026-09-21T19:00:55.137007+00:00", "framework": "vllm", "intentId": "d6229ef4ac9942078baa35e312133d9d", "repoId": "neuralmagic/Llama-3.2-1B-Instruct-quantized.w8a8", "safeConfigVector": {"dtype": "-", "gpuNum": 1, "maxModelLen": 2048}, "status": "age_policy_deferred", "targetGpu": "Ascend_910-b4", "taskId": null, "taskType": "text-generation"} diff --git a/manifest.json b/manifest.json index 197fee334f..1573a41245 100644 --- a/manifest.json +++ b/manifest.json @@ -2,24 +2,24 @@ "agentVersion": "2026.09.20.2", "checksums": { ".modelhub_state/account_capacity.json": "930f9869793c3d1aa071c53f05b7cf82a4e372f002be88361d6d6189e27c12a1", - ".modelhub_state/architecture_compatibility_blacklist.json": "867c758f46b25856a39ff32773717a68f61e6753c678b8953ad9956a86622009", + ".modelhub_state/architecture_compatibility_blacklist.json": "0cb3fa906e6010e368797f6d0659f3358e3f921784e6c67507274af25a8a405c", ".modelhub_state/architecture_history_backfill.json": "a92348206605da04f75c9e26e7c818b76f259e0b28983f18b6375ef94802f0ab", - ".modelhub_state/market_intelligence.json": "e24d68b4d1d42e299627d88c563c3d6eec953077a4037747dbadfda48293b7c6", - ".modelhub_state/official_capabilities.json": "1a8b45609856ad20ebd86a96ada4f48b0c99dc3167f6bd56f190f7ba9c316c29", - ".modelhub_state/outcome_checkpoint.json": "f395863f9340925d3d9933b0e90168af86d3752c1bc496d4a7bc6794856e2783", + ".modelhub_state/market_intelligence.json": "049787af8f22e4597c647bff59a35764b31a3ab1be6ec08a46ebe3810d0a22b1", + ".modelhub_state/official_capabilities.json": "de07455fa51772e4f4122355d2b912cf7f944ae1f881af9285cd2d4b1532d00e", + ".modelhub_state/outcome_checkpoint.json": "594c9d2c72b0572744c9fd3adb402a07bc0c02d11564ff6beacfdf7cac849935", ".modelhub_state/queue_cleanup_latest.json": "14282fb49e9aa62e07b6a07250bcf51916a4d783e09664c80d1fa965b4d48615", ".modelhub_state/recent_outcomes.jsonl": "b2ab7ac8f4a931c0e508f2b4c3073fa64ccc6cbdc51d88abc863625e9f7a5ae2", ".modelhub_state/recovery_active_tasks.jsonl": "5d4c5c1f4e4d56f7f16e855e781e9e08e22b1e8ae6c31c3056cab3f9ef5bf7ea", - ".modelhub_state/recovery_intents.jsonl": "31dfeb716b04774a930c5b3c82e763de46601606e4f2bad234de53aa0d796c5b", + ".modelhub_state/recovery_intents.jsonl": "3036b9c9583e62727d02594b259fe6ee53462f8faa2b43026c623e8eae821426", ".modelhub_state/routing_intelligence.json": "825abf0ace473b3ac58457291c5880c0e55ff12a20c7d5bd8bbea2a5d559b101", ".modelhub_state/submission_exclusions.jsonl": "80315f370e9cc3cdadaf390e12573ef887794214779379c50b7b37b7631e8989", ".modelhub_state/worker_crashes.jsonl": "07505b69e0b050da5f90f8b046a3069d3e1ca2574ca39896bc7d8a66293167f7", "ledger/submissions.jsonl": "3ab89cecd55636c6f932db7311d37b66ea7555fedd906a11f1731e131b7867e8", - "outcomes/submissions.jsonl": "04ed0b85d91a15089be96e6cedb6ecaca97670d4bd9f30dedda04f573acf9ac8" + "outcomes/submissions.jsonl": "f532ede22ba0f924711edd318286e98d1683b1a46a28ca753bdee9aa6aa3b8c9" }, - "generation": 11938, - "phase": "cycle_no_slots", + "generation": 11939, + "phase": "intent", "schemaVersion": 1, - "updatedAt": "2026-09-21T23:59:18.690363+00:00", + "updatedAt": "2026-09-22T00:00:42.907898+00:00", "writerId": "8b35139af6674067a339a670222d4b67" } diff --git a/outcomes/submissions.jsonl b/outcomes/submissions.jsonl index 5ee0f51bc0..a8feeec9ba 100644 --- a/outcomes/submissions.jsonl +++ b/outcomes/submissions.jsonl @@ -491,7 +491,6 @@ {"failReason": null, "framework": "vllm", "lastSyncTime": "2026-09-20T12:01:00.671063+00:00", "modelId": "aisingapore/SEA-LION-v1-7B-IT-Research", "modelProfile": {"architectures": ["MPTForCausalLM"], "configFingerprint": "e4f10af7e21b1a730b2dfa32856745ab0e7a14867778026d61065fda3de90727", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 15003361968, "estimatedRequiredGiB": 16.773, "gpuMemoryEvidence": {"memoryGiB": 64.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 64.0, "maximumRepositorySizeGiB": 53.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "mpt", "modelscopeFileSize": 15008118474, "modelscopeLicense": "cc-by-nc-sa-4.0", "modelscopeParams": 7501651968, "modelscopeTags": ["license:cc-by-nc-sa-4.0", "model_type:mpt", "library:transformer", "library:safetensors", "library:pytorch", "task:text-generation"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 15008118474}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-20T03:41:49.739370+00:00", "targetGpu": "hygon_k100-ai", "taskId": "4970046", "taskType": "text-generation", "verifyResult": null} {"failReason": null, "framework": "vllm_tokenizer_patch", "lastSyncTime": "2026-09-20T12:01:00.670485+00:00", "modelId": "OpenBMB/MiniCPM5-2B-MLX", "modelProfile": {"architectures": ["LlamaForCausalLM"], "configFingerprint": "1c0a4f69628b2a0ad37a6eee139a4c98afed7d8c93529a9ef42a90756e4e1940", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 1416035216, "estimatedRequiredGiB": 1.594, "gpuMemoryEvidence": {"memoryGiB": 100.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 100.0, "maximumRepositorySizeGiB": 83.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "llama", "modelscopeFileSize": 1426233716, "modelscopeLicense": "apache-2.0", "modelscopeParams": 393390080, "modelscopeTags": ["license:apache-2.0", "model_type:llama", "library:transformer", "library:safetensors", "task:text-generation", "custom_tag:minicpm", "custom_tag:minicpm5", "custom_tag:llama", "custom_tag:text-generation", "custom_tag:long-context", "custom_tag:tool-calling", "custom_tag:on-device", "custom_tag:edge-ai"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 1426233716}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-20T03:46:29.638998+00:00", "targetGpu": "Kunlunxin_p-800", "taskId": "4970108", "taskType": "text-generation", "verifyResult": null} {"failReason": null, "framework": "vllm_tokenizer_patch", "lastSyncTime": "2026-09-20T12:01:00.671117+00:00", "modelId": "inclusionAI/Ling-3.0-tiny", "modelProfile": {"architectures": ["BailingMoeV3ForCausalLM"], "configFingerprint": "1c0a4f69628b2a0ad37a6eee139a4c98afed7d8c93529a9ef42a90756e4e1940", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 15787992416, "estimatedRequiredGiB": 17.659, "gpuMemoryEvidence": {"memoryGiB": 100.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 100.0, "maximumRepositorySizeGiB": 83.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "bailing_hybrid", "modelscopeFileSize": 15801179822, "modelscopeLicense": "mit", "modelscopeParams": 7893392800, "modelscopeTags": ["license:mit", "model_type:bailing_hybrid", "library:safetensors", "library:pytorch", "task:text-generation"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 15801179822}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-20T03:46:29.642501+00:00", "targetGpu": "Kunlunxin_p-800", "taskId": "4970106", "taskType": "text-generation", "verifyResult": null} -{"failReason": null, "framework": "vllm_tokenizer_patch", "lastSyncTime": "2026-09-20T12:01:00.671197+00:00", "modelId": "solidrust/Llama-3-8B-Instruct-DPO-v0.1-AWQ", "modelProfile": {"architectures": ["LlamaForCausalLM"], "configFingerprint": "1c0a4f69628b2a0ad37a6eee139a4c98afed7d8c93529a9ef42a90756e4e1940", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 5727954960, "estimatedRequiredGiB": 6.412, "gpuMemoryEvidence": {"memoryGiB": 100.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 100.0, "maximumRepositorySizeGiB": 83.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "llama", "modelscopeFileSize": 5737317601, "modelscopeLicense": "other", "modelscopeParams": 8030269440, "modelscopeTags": ["license:other", "model_type:llama", "library:transformer", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:4-bit", "custom_tag:AWQ", "custom_tag:text-generation", "custom_tag:autotrain_compatible", "custom_tag:endpoints_compatible", "custom_tag:axolotl", "custom_tag:finetune", "custom_tag:facebook", "custom_tag:meta", "custom_tag:pytorch", "custom_tag:llama", "custom_tag:llama-3"], "modelscopeTasks": ["text-generation"], "quantizationMethod": "awq", "repositoryOnDiskBytes": 5737317601}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-20T03:46:29.640961+00:00", "targetGpu": "Kunlunxin_p-800", "taskId": "4970105", "taskType": "text-generation", "verifyResult": null} {"failReason": null, "framework": "vllm_tokenizer_patch", "lastSyncTime": "2026-09-20T12:01:00.670681+00:00", "modelId": "solidrust/Llama-3-8B-Instruct-v0.5-AWQ", "modelProfile": {"architectures": ["LlamaForCausalLM"], "configFingerprint": "1c0a4f69628b2a0ad37a6eee139a4c98afed7d8c93529a9ef42a90756e4e1940", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 5727954960, "estimatedRequiredGiB": 6.412, "gpuMemoryEvidence": {"memoryGiB": 100.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 100.0, "maximumRepositorySizeGiB": 83.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "llama", "modelscopeFileSize": 5737318602, "modelscopeLicense": "other", "modelscopeParams": 8030269440, "modelscopeTags": ["license:other", "model_type:llama", "library:transformer", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:4-bit", "custom_tag:AWQ", "custom_tag:text-generation", "custom_tag:autotrain_compatible", "custom_tag:endpoints_compatible", "custom_tag:axolotl", "custom_tag:finetune", "custom_tag:facebook", "custom_tag:meta", "custom_tag:pytorch", "custom_tag:llama", "custom_tag:llama-3"], "modelscopeTasks": ["text-generation"], "quantizationMethod": "awq", "repositoryOnDiskBytes": 5737318602}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-20T03:46:29.646657+00:00", "targetGpu": "Kunlunxin_p-800", "taskId": "4970109", "taskType": "text-generation", "verifyResult": null} {"failReason": null, "framework": "vllm_tokenizer_patch", "lastSyncTime": "2026-09-20T12:01:00.670754+00:00", "modelId": "solidrust/Llama-3-11B-Instruct-v0.1-AWQ", "modelProfile": {"architectures": ["LlamaForCausalLM"], "configFingerprint": "1c0a4f69628b2a0ad37a6eee139a4c98afed7d8c93529a9ef42a90756e4e1940", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 7541230432, "estimatedRequiredGiB": 8.438, "gpuMemoryEvidence": {"memoryGiB": 100.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 100.0, "maximumRepositorySizeGiB": 83.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "llama", "modelscopeFileSize": 7550622334, "modelscopeLicense": "other", "modelscopeParams": 11520053248, "modelscopeTags": ["license:other", "model_type:llama", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:4-bit", "custom_tag:AWQ", "custom_tag:text-generation", "custom_tag:autotrain_compatible", "custom_tag:endpoints_compatible", "custom_tag:mergekit", "custom_tag:merge", "custom_tag:facebook", "custom_tag:meta", "custom_tag:pytorch", "custom_tag:llama", "custom_tag:llama-3"], "modelscopeTasks": ["text-generation"], "quantizationMethod": "awq", "repositoryOnDiskBytes": 7550622334}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-20T03:46:29.645217+00:00", "targetGpu": "Kunlunxin_p-800", "taskId": "4970116", "taskType": "text-generation", "verifyResult": null} {"failReason": null, "framework": "vllm_tokenizer_patch", "lastSyncTime": "2026-09-20T12:01:00.671265+00:00", "modelId": "solidrust/Llama-3-13B-Instruct-v0.1-AWQ", "modelProfile": {"architectures": ["LlamaForCausalLM"], "configFingerprint": "1c0a4f69628b2a0ad37a6eee139a4c98afed7d8c93529a9ef42a90756e4e1940", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 8447876488, "estimatedRequiredGiB": 9.452, "gpuMemoryEvidence": {"memoryGiB": 100.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 100.0, "maximumRepositorySizeGiB": 83.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "llama", "modelscopeFileSize": 8457284244, "modelscopeLicense": "other", "modelscopeParams": 13264949248, "modelscopeTags": ["license:other", "model_type:llama", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:4-bit", "custom_tag:AWQ", "custom_tag:text-generation", "custom_tag:autotrain_compatible", "custom_tag:endpoints_compatible", "custom_tag:mergekit", "custom_tag:merge", "custom_tag:facebook", "custom_tag:meta", "custom_tag:pytorch", "custom_tag:llama", "custom_tag:llama-3"], "modelscopeTasks": ["text-generation"], "quantizationMethod": "awq", "repositoryOnDiskBytes": 8457284244}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-20T03:46:29.594810+00:00", "targetGpu": "Kunlunxin_p-800", "taskId": "4970112", "taskType": "text-generation", "verifyResult": null}