CoolBoy/submmit
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

state: generation 5986 (cycle)

Browse Source
This commit is contained in:
CoolBoy
2026-09-12 02:41:14 +00:00
parent 5a5a676610
commit 29502df052
10 changed files with 2203 additions and 2186 deletions
Download Patch File Download Diff File Expand all files Collapse all files

129
.modelhub_state/outcome_checkpoint.json
View File

@@ -1,6 +1,6 @@
{
"generatedAt": "2026-09-12T02:30:20.660051+00:00",
"lastSyncTime": "2026-09-12T02:30:20.417896+00:00",
"generatedAt": "2026-09-12T02:39:29.471961+00:00",
"lastSyncTime": "2026-09-12T02:39:29.219629+00:00",
"recentLimit": 300,
"report": {
"architectureCompatibilityBlocks": {
@@ -992,6 +992,25 @@
"targetGpu": "Mthreads_s4000",
"taskType": "text-generation"
},
"sunrise_pt-200-x1|vllm_fix_tokenizer|text-generation|model_type:mellum": {
"architectureSignature": "model_type:mellum",
"architectures": [],
"evidenceCount": 1,
"expiresAt": "2026-10-12T02:35:21+00:00",
"framework": "vllm_fix_tokenizer",
"latestFailureAt": "2026-09-12T02:35:21+00:00",
"latestSuccessfulAt": null,
"matchType": "model_type",
"modelType": "mellum",
"sourceModelIds": [
"JetBrains/Mellum2-12B-A2.5B-Thinking"
],
"sourceTaskIds": [
"4152611"
],
"targetGpu": "Sunrise_pt-200-x1",
"taskType": "text-generation"
},
"sunrise_pt-200-x1|vllm_fix_tokenizer|text-generation|model_type:qwen3_5": {
"architectureSignature": "model_type:qwen3_5",
"architectures": [],
@@ -1312,7 +1331,7 @@
}
},
"architectureCompatibilitySummary": {
"activeBlockCount": 66,
"activeBlockCount": 67,
"byGpuFramework": {
"Ascend_910-b3|vllm": 8,
"Ascend_910-b4|vllm": 9,
@@ -1322,7 +1341,7 @@
"MetaX_c-500|vllm": 7,
"Mthreads_s4000|vllm": 7,
"Sunrise_pt-200-x1|vllm": 1,
"Sunrise_pt-200-x1|vllm_fix_tokenizer": 1,
"Sunrise_pt-200-x1|vllm_fix_tokenizer": 2,
"Vastai_va16|vllm": 12,
"Vastai_va16|vllm_fix_tokenizer": 2,
"hygon_k100-ai|vllm": 7,
@@ -2063,17 +2082,17 @@
"unresolvedFailureCount": 33
},
"Sunrise_pt-200-x1|vllm_fix_tokenizer|text-generation": {
"attributableFailureCount": 37,
"attributableFailureCount": 39,
"decisionFailureRate": 1.0,
"decisionSuccessRate": 0.0,
"decisionTotal": 37,
"decisionTotal": 39,
"failureBreakdown": {
"framework_architecture_unsupported": 2,
"framework_architecture_unsupported": 3,
"platform_infrastructure": 2,
"tokenizer_compatibility": 35,
"tokenizer_compatibility": 36,
"参数/模板问题": 6
},
"failureCount": 45,
"failureCount": 47,
"failureRate": 1.0,
"framework": "vllm_fix_tokenizer",
"pendingCount": 0,
@@ -2083,7 +2102,7 @@
"successRate": 0.0,
"targetGpu": "Sunrise_pt-200-x1",
"taskType": "text-generation",
"total": 45,
"total": 47,
"unresolvedFailureCount": 6
},
"Sunrise_pt-200-x1|vllm|text-generation": {
@@ -2423,26 +2442,26 @@
"unresolvedFailureCount": 4
},
"vllm_fix_tokenizer": {
"attributableFailureCount": 40,
"attributableFailureCount": 42,
"decisionFailureRate": 1.0,
"decisionSuccessRate": 0.0,
"decisionTotal": 40,
"decisionTotal": 42,
"failureBreakdown": {
"ambiguous_runtime": 36,
"framework_architecture_unsupported": 4,
"framework_architecture_unsupported": 5,
"model_load": 1,
"platform_infrastructure": 2,
"tokenizer_compatibility": 35,
"tokenizer_compatibility": 36,
"参数/模板问题": 7
},
"failureCount": 85,
"failureCount": 87,
"failureRate": 1.0,
"pendingCount": 0,
"pendingRate": 0.0,
"platformFailureCount": 2,
"successCount": 0,
"successRate": 0.0,
"total": 85,
"total": 87,
"unresolvedFailureCount": 43
},
"vllm_tokenizer_patch": {
@@ -2464,7 +2483,7 @@
"unresolvedFailureCount": 1
}
},
"generatedAt": "2026-09-12T02:30:20.655640+00:00",
"generatedAt": "2026-09-12T02:39:29.466891+00:00",
"gpuSummaries": {
"Ascend_910-b3": {
"attributableFailureCount": 26,
@@ -2722,26 +2741,26 @@
"unresolvedFailureCount": 77
},
"Sunrise_pt-200-x1": {
"attributableFailureCount": 39,
"decisionFailureRate": 0.975,
"decisionSuccessRate": 0.025,
"decisionTotal": 40,
"attributableFailureCount": 41,
"decisionFailureRate": 0.9762,
"decisionSuccessRate": 0.0238,
"decisionTotal": 42,
"failureBreakdown": {
"ambiguous_runtime": 1,
"framework_architecture_unsupported": 3,
"framework_architecture_unsupported": 4,
"platform_infrastructure": 3,
"tokenizer_compatibility": 36,
"tokenizer_compatibility": 37,
"参数/模板问题": 9,
"验证失败": 32
},
"failureCount": 84,
"failureRate": 0.9882,
"failureCount": 86,
"failureRate": 0.9885,
"pendingCount": 0,
"pendingRate": 0.0,
"platformFailureCount": 3,
"successCount": 1,
"successRate": 0.0118,
"total": 85,
"successRate": 0.0115,
"total": 87,
"unresolvedFailureCount": 42
},
"Vastai_va16": {
@@ -4303,19 +4322,19 @@
"unresolvedFailureCount": 1
},
"Iluvatar_bi-150|vllm|text-generation": {
"attributableFailureCount": 5,
"consecutiveFailures": 5,
"attributableFailureCount": 4,
"consecutiveFailures": 4,
"consecutivePlatformFailures": 0,
"decisionFailureRate": 1.0,
"decisionSuccessRate": 0.0,
"decisionTotal": 5,
"decisionTotal": 4,
"failureBreakdown": {
"ambiguous_runtime": 2,
"backend_operator": 1,
"framework_architecture_unsupported": 3,
"framework_architecture_unsupported": 2,
"model_load": 1
},
"failureCount": 7,
"failureCount": 6,
"failureRate": 1.0,
"framework": "vllm",
"lastPlatformFailureAt": null,
@@ -4327,7 +4346,7 @@
"successRate": 0.0,
"targetGpu": "Iluvatar_bi-150",
"taskType": "text-generation",
"total": 7,
"total": 6,
"unresolvedFailureCount": 2
},
"Iluvatar_mrv-100|unknown|text-generation": {
@@ -4509,22 +4528,22 @@
"unresolvedFailureCount": 0
},
"Sunrise_pt-200-x1|vllm_fix_tokenizer|text-generation": {
"attributableFailureCount": 16,
"consecutiveFailures": 16,
"attributableFailureCount": 18,
"consecutiveFailures": 18,
"consecutivePlatformFailures": 0,
"decisionFailureRate": 1.0,
"decisionSuccessRate": 0.0,
"decisionTotal": 16,
"decisionTotal": 18,
"failureBreakdown": {
"framework_architecture_unsupported": 1,
"framework_architecture_unsupported": 2,
"platform_infrastructure": 2,
"tokenizer_compatibility": 15
"tokenizer_compatibility": 16
},
"failureCount": 18,
"failureCount": 20,
"failureRate": 1.0,
"framework": "vllm_fix_tokenizer",
"lastPlatformFailureAt": "2026-09-10T11:01:21+00:00",
"lastTerminalAt": "2026-09-11T11:19:08.306628+00:00",
"lastTerminalAt": "2026-09-12T02:39:29.219599+00:00",
"pendingCount": 0,
"pendingRate": 0.0,
"platformFailureCount": 2,
@@ -4532,7 +4551,7 @@
"successRate": 0.0,
"targetGpu": "Sunrise_pt-200-x1",
"taskType": "text-generation",
"total": 18,
"total": 20,
"unresolvedFailureCount": 0
},
"Sunrise_pt-200-x1|vllm|text-generation": {
@@ -7055,41 +7074,40 @@
"unresolvedFailureCount": 0
}
},
"terminalRecords": 1673,
"totalRecords": 1760,
"terminalRecords": 1675,
"totalRecords": 1762,
"totals": {
"attributableFailureCount": 445,
"decisionFailureRate": 0.8882,
"decisionSuccessRate": 0.1118,
"decisionTotal": 501,
"attributableFailureCount": 447,
"decisionFailureRate": 0.8887,
"decisionSuccessRate": 0.1113,
"decisionTotal": 503,
"failureBreakdown": {
"ambiguous_runtime": 394,
"backend_operator": 40,
"framework_architecture_unsupported": 277,
"framework_architecture_unsupported": 278,
"memory_capacity": 11,
"model_load": 64,
"platform_infrastructure": 4,
"repository_structure": 5,
"runtime_memory": 6,
"tokenizer_compatibility": 42,
"tokenizer_compatibility": 43,
"参数/模板问题": 101,
"验证失败": 673
},
"failureCount": 1617,
"failureRate": 0.9665,
"failureCount": 1619,
"failureRate": 0.9666,
"pendingCount": 0,
"pendingRate": 0.0,
"platformFailureCount": 4,
"successCount": 56,
"successRate": 0.0335,
"total": 1673,
"successRate": 0.0334,
"total": 1675,
"unresolvedFailureCount": 1168
},
"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。",
@@ -7098,6 +7116,7 @@
"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。",
@@ -7118,6 +7137,6 @@
]
},
"storageMode": "decision_state_only",
"summarizedRecords": 1760,
"summarizedRecords": 1762,
"version": 1
}