state: generation 6428 (cycle)

This commit is contained in:
2026-09-13 00:12:36 +00:00
parent c0449ea4c6
commit 66932b8bed
11 changed files with 2160 additions and 2148 deletions

View File

@@ -1,6 +1,6 @@
{
"generatedAt": "2026-09-12T23:53:22.974979+00:00",
"lastSyncTime": "2026-09-12T23:53:22.743813+00:00",
"generatedAt": "2026-09-13T00:10:37.920504+00:00",
"lastSyncTime": "2026-09-13T00:10:37.716599+00:00",
"recentLimit": 300,
"report": {
"architectureCompatibilityBlocks": {
@@ -558,6 +558,25 @@
"targetGpu": "hygon_k100-ai",
"taskType": "text-generation"
},
"hygon_k100-ai|vllm|text-generation|model_type:gemma4": {
"architectureSignature": "model_type:gemma4",
"architectures": [],
"evidenceCount": 1,
"expiresAt": "2026-10-13T00:07:21+00:00",
"framework": "vllm",
"latestFailureAt": "2026-09-13T00:07:21+00:00",
"latestSuccessfulAt": null,
"matchType": "model_type",
"modelType": "gemma4",
"sourceModelIds": [
"nv-community/Gemma-4-31B-IT-NVFP4"
],
"sourceTaskIds": [
"4079956"
],
"targetGpu": "hygon_k100-ai",
"taskType": "text-generation"
},
"hygon_k100-ai|vllm|text-generation|model_type:mellum": {
"architectureSignature": "model_type:mellum",
"architectures": [],
@@ -1436,7 +1455,7 @@
}
},
"architectureCompatibilitySummary": {
"activeBlockCount": 72,
"activeBlockCount": 73,
"byGpuFramework": {
"Ascend_910-b3|vllm": 9,
"Ascend_910-b4|vllm": 10,
@@ -1449,7 +1468,7 @@
"Sunrise_pt-200-x1|vllm_fix_tokenizer": 3,
"Vastai_va16|vllm": 13,
"Vastai_va16|vllm_fix_tokenizer": 2,
"hygon_k100-ai|vllm": 7,
"hygon_k100-ai|vllm": 8,
"hygon_k100-ai|vllm-patch-tokenizer": 2
},
"ttlDays": 30
@@ -2374,19 +2393,19 @@
"unresolvedFailureCount": 8
},
"hygon_k100-ai|vllm|text-generation": {
"attributableFailureCount": 42,
"attributableFailureCount": 43,
"decisionFailureRate": 1.0,
"decisionSuccessRate": 0.0,
"decisionTotal": 42,
"decisionTotal": 43,
"failureBreakdown": {
"ambiguous_runtime": 18,
"framework_architecture_unsupported": 31,
"framework_architecture_unsupported": 32,
"memory_capacity": 1,
"model_load": 4,
"repository_structure": 4,
"runtime_memory": 2
},
"failureCount": 60,
"failureCount": 61,
"failureRate": 1.0,
"framework": "vllm",
"pendingCount": 0,
@@ -2396,7 +2415,7 @@
"successRate": 0.0,
"targetGpu": "hygon_k100-ai",
"taskType": "text-generation",
"total": 60,
"total": 61,
"unresolvedFailureCount": 18
}
},
@@ -2465,14 +2484,14 @@
"unresolvedFailureCount": 870
},
"vllm": {
"attributableFailureCount": 283,
"attributableFailureCount": 284,
"decisionFailureRate": 0.9895,
"decisionSuccessRate": 0.0105,
"decisionTotal": 286,
"decisionTotal": 287,
"failureBreakdown": {
"ambiguous_runtime": 201,
"backend_operator": 34,
"framework_architecture_unsupported": 197,
"framework_architecture_unsupported": 198,
"memory_capacity": 6,
"model_load": 26,
"platform_infrastructure": 1,
@@ -2481,14 +2500,14 @@
"tokenizer_compatibility": 6,
"参数/模板问题": 24
},
"failureCount": 509,
"failureRate": 0.9941,
"failureCount": 510,
"failureRate": 0.9942,
"pendingCount": 0,
"pendingRate": 0.0,
"platformFailureCount": 1,
"successCount": 3,
"successRate": 0.0059,
"total": 512,
"successRate": 0.0058,
"total": 513,
"unresolvedFailureCount": 225
},
"vllm-mlu": {
@@ -2593,7 +2612,7 @@
"unresolvedFailureCount": 1
}
},
"generatedAt": "2026-09-12T23:53:22.970900+00:00",
"generatedAt": "2026-09-13T00:10:37.916223+00:00",
"gpuSummaries": {
"Ascend_910-b3": {
"attributableFailureCount": 30,
@@ -2900,13 +2919,13 @@
"unresolvedFailureCount": 206
},
"hygon_k100-ai": {
"attributableFailureCount": 46,
"attributableFailureCount": 47,
"decisionFailureRate": 1.0,
"decisionSuccessRate": 0.0,
"decisionTotal": 46,
"decisionTotal": 47,
"failureBreakdown": {
"ambiguous_runtime": 20,
"framework_architecture_unsupported": 35,
"framework_architecture_unsupported": 36,
"memory_capacity": 1,
"model_load": 4,
"repository_structure": 4,
@@ -2914,14 +2933,14 @@
"参数/模板问题": 9,
"验证失败": 27
},
"failureCount": 102,
"failureCount": 103,
"failureRate": 1.0,
"pendingCount": 0,
"pendingRate": 0.0,
"platformFailureCount": 0,
"successCount": 0,
"successRate": 0.0,
"total": 102,
"total": 103,
"unresolvedFailureCount": 56
}
},
@@ -4914,15 +4933,15 @@
"unresolvedFailureCount": 1
},
"hygon_k100-ai|vllm|text-generation": {
"attributableFailureCount": 7,
"consecutiveFailures": 7,
"attributableFailureCount": 8,
"consecutiveFailures": 8,
"consecutivePlatformFailures": 0,
"decisionFailureRate": 1.0,
"decisionSuccessRate": 0.0,
"decisionTotal": 7,
"decisionTotal": 8,
"failureBreakdown": {
"ambiguous_runtime": 13,
"framework_architecture_unsupported": 4,
"ambiguous_runtime": 12,
"framework_architecture_unsupported": 5,
"model_load": 1,
"repository_structure": 1,
"runtime_memory": 1
@@ -4931,7 +4950,7 @@
"failureRate": 1.0,
"framework": "vllm",
"lastPlatformFailureAt": null,
"lastTerminalAt": "2026-09-12T11:23:23.155704+00:00",
"lastTerminalAt": "2026-09-13T00:10:37.716599+00:00",
"pendingCount": 0,
"pendingRate": 0.0,
"platformFailureCount": 0,
@@ -4940,7 +4959,7 @@
"targetGpu": "hygon_k100-ai",
"taskType": "text-generation",
"total": 20,
"unresolvedFailureCount": 13
"unresolvedFailureCount": 12
}
},
"recentProfileCombinationStats": {
@@ -7273,17 +7292,17 @@
"unresolvedFailureCount": 0
}
},
"terminalRecords": 1745,
"totalRecords": 1832,
"terminalRecords": 1746,
"totalRecords": 1833,
"totals": {
"attributableFailureCount": 498,
"decisionFailureRate": 0.8957,
"decisionSuccessRate": 0.1043,
"decisionTotal": 556,
"attributableFailureCount": 499,
"decisionFailureRate": 0.8959,
"decisionSuccessRate": 0.1041,
"decisionTotal": 557,
"failureBreakdown": {
"ambiguous_runtime": 411,
"backend_operator": 42,
"framework_architecture_unsupported": 310,
"framework_architecture_unsupported": 311,
"memory_capacity": 11,
"model_load": 73,
"platform_infrastructure": 4,
@@ -7293,20 +7312,21 @@
"参数/模板问题": 101,
"验证失败": 673
},
"failureCount": 1687,
"failureCount": 1688,
"failureRate": 0.9668,
"pendingCount": 0,
"pendingRate": 0.0,
"platformFailureCount": 4,
"successCount": 58,
"successRate": 0.0332,
"total": 1745,
"total": 1746,
"unresolvedFailureCount": 1185
},
"warnings": [
"GPU Ascend_910-b3 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 Mthreads_s4000 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 hygon_k100-ai 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_mrv-100 本地统计失败率偏高≥50%),建议重点关注。",
@@ -7315,7 +7335,6 @@
"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-150 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。",
"组合 hygon_k100-ai|vllm-patch-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 hygon_k100-ai|vllm|text-generation 近期失败集中,建议降低该 GPU+框架的提交优先级。",
"组合 Ascend_910-b4|vllm|text-generation 近期失败集中,建议降低该 GPU+框架的提交优先级。",
@@ -7338,6 +7357,6 @@
]
},
"storageMode": "decision_state_only",
"summarizedRecords": 1832,
"summarizedRecords": 1833,
"version": 1
}