state: generation 23023 (cycle)

.modelhub_state/architecture_compatibility_blacklist.json

@@ -2965,7 +2965,7 @@
       "taskType": "text-generation"
     }
   },
-  "generatedAt": "2026-10-10T14:40:22.764557+00:00",
+  "generatedAt": "2026-10-10T14:53:24.261634+00:00",
   "summary": {
     "activeBlockCount": 148,
     "byGpuFramework": {

.modelhub_state/market_intelligence.json

@@ -443,7 +443,7 @@
     }
   },
   "frameworkUpdatedAt": null,
-  "generatedAt": "2026-10-10T14:51:29.070459+00:00",
+  "generatedAt": "2026-10-10T14:55:00.672949+00:00",
   "gpuStats": {
     "Ascend_910-b3": {
       "available": true,

.modelhub_state/outcome_checkpoint.json

@@ -1,6 +1,6 @@
 {
-  "generatedAt": "2026-10-10T13:57:15.330792+00:00",
+  "generatedAt": "2026-10-10T14:53:24.178346+00:00",
-  "lastSyncTime": "2026-10-10T13:57:15.079577+00:00",
+  "lastSyncTime": "2026-10-10T14:53:23.961113+00:00",
   "recentLimit": 300,
   "report": {
     "architectureCompatibilityBlocks": {
@@ -4639,10 +4639,10 @@
         "decisionSuccessRate": 0.0,
         "decisionTotal": 0,
         "failureBreakdown": {
-          "ambiguous_runtime": 185,
+          "ambiguous_runtime": 186,
           "参数/模板问题": 4
         },
-        "failureCount": 189,
+        "failureCount": 190,
         "failureRate": 1.0,
         "framework": "vllm_fix_tokenizer",
         "pendingCount": 0,
@@ -4652,8 +4652,8 @@
         "successRate": 0.0,
         "targetGpu": "Kunlunxin_p-800",
         "taskType": "text-generation",
-        "total": 189,
+        "total": 190,
-        "unresolvedFailureCount": 189
+        "unresolvedFailureCount": 190
       },
       "Kunlunxin_p-800|vllm_tokenizer_patch|text-generation": {
         "attributableFailureCount": 61,
@@ -5826,7 +5826,7 @@
         "decisionSuccessRate": 0.0269,
         "decisionTotal": 186,
         "failureBreakdown": {
-          "ambiguous_runtime": 415,
+          "ambiguous_runtime": 416,
           "backend_operator": 10,
           "framework_architecture_unsupported": 52,
           "memory_capacity": 10,
@@ -5835,15 +5835,15 @@
           "tokenizer_compatibility": 96,
           "参数/模板问题": 16
         },
-        "failureCount": 639,
+        "failureCount": 640,
         "failureRate": 0.9922,
         "pendingCount": 0,
         "pendingRate": 0.0,
         "platformFailureCount": 27,
         "successCount": 5,
         "successRate": 0.0078,
-        "total": 644,
+        "total": 645,
-        "unresolvedFailureCount": 431
+        "unresolvedFailureCount": 432
       },
       "vllm_tokenizer_patch": {
         "attributableFailureCount": 170,
@@ -5874,7 +5874,7 @@
         "unresolvedFailureCount": 323
       }
     },
-    "generatedAt": "2026-10-10T13:57:15.318133+00:00",
+    "generatedAt": "2026-10-10T14:53:24.165449+00:00",
     "gpuSummaries": {
       "Ascend_910-b3": {
         "attributableFailureCount": 130,
@@ -6106,7 +6106,7 @@
         "decisionSuccessRate": 0.066,
         "decisionTotal": 106,
         "failureBreakdown": {
-          "ambiguous_runtime": 274,
+          "ambiguous_runtime": 275,
           "backend_operator": 23,
           "framework_architecture_unsupported": 37,
           "memory_capacity": 3,
@@ -6119,15 +6119,15 @@
           "日志缺失": 1,
           "验证失败": 23
         },
-        "failureCount": 446,
+        "failureCount": 447,
-        "failureRate": 0.9845,
+        "failureRate": 0.9846,
         "pendingCount": 0,
         "pendingRate": 0.0,
         "platformFailureCount": 1,
         "successCount": 7,
-        "successRate": 0.0155,
+        "successRate": 0.0154,
-        "total": 453,
+        "total": 454,
-        "unresolvedFailureCount": 346
+        "unresolvedFailureCount": 347
       },
       "Kunlunxin_r-200-8f": {
         "attributableFailureCount": 0,
@@ -19124,9 +19124,9 @@
         "decisionSuccessRate": 0.0,
         "decisionTotal": 0,
         "failureBreakdown": {
-          "ambiguous_runtime": 15
+          "ambiguous_runtime": 16
         },
-        "failureCount": 15,
+        "failureCount": 16,
         "failureRate": 1.0,
         "framework": "vllm_fix_tokenizer",
         "modelType": "mini_k3",
@@ -19138,8 +19138,8 @@
         "successRate": 0.0,
         "targetGpu": "Kunlunxin_p-800",
         "taskType": "text-generation",
-        "total": 15,
+        "total": 16,
-        "unresolvedFailureCount": 15
+        "unresolvedFailureCount": 16
       },
       "Kunlunxin_p-800|vllm_fix_tokenizer|text-generation|minicpm|none": {
         "attributableFailureCount": 0,
@@ -47669,9 +47669,9 @@
         "decisionSuccessRate": 0.0,
         "decisionTotal": 0,
         "failureBreakdown": {
-          "ambiguous_runtime": 15
+          "ambiguous_runtime": 16
         },
-        "failureCount": 15,
+        "failureCount": 16,
         "failureRate": 1.0,
         "framework": "vllm_fix_tokenizer",
         "loadSizeLog2Bucket": 30,
@@ -47684,8 +47684,8 @@
         "successRate": 0.0,
         "targetGpu": "Kunlunxin_p-800",
         "taskType": "text-generation",
-        "total": 15,
+        "total": 16,
-        "unresolvedFailureCount": 15
+        "unresolvedFailureCount": 16
       },
       "Kunlunxin_p-800|vllm_fix_tokenizer|text-generation|minicpm|none|33": {
         "attributableFailureCount": 0,
@@ -58686,15 +58686,15 @@
         "unresolvedFailureCount": 0
       }
     },
-    "terminalRecords": 17807,
+    "terminalRecords": 17808,
-    "totalRecords": 18325,
+    "totalRecords": 18326,
     "totals": {
       "attributableFailureCount": 6316,
       "decisionFailureRate": 0.8656,
       "decisionSuccessRate": 0.1344,
       "decisionTotal": 7297,
       "failureBreakdown": {
-        "ambiguous_runtime": 4631,
+        "ambiguous_runtime": 4632,
         "architecture_compatibility": 212,
         "attention_backend": 3,
         "backend_operator": 131,
@@ -58710,15 +58710,15 @@
         "日志缺失": 719,
         "验证失败": 676
       },
-      "failureCount": 16826,
+      "failureCount": 16827,
       "failureRate": 0.9449,
       "pendingCount": 0,
       "pendingRate": 0.0,
       "platformFailureCount": 956,
       "successCount": 981,
       "successRate": 0.0551,
-      "total": 17807,
+      "total": 17808,
-      "unresolvedFailureCount": 9554
+      "unresolvedFailureCount": 9555
     },
     "warnings": [
       "GPU MetaX_c-500 本地统计失败率偏高（≥50%），建议重点关注。",
@@ -58780,6 +58780,6 @@
     ]
   },
   "storageMode": "decision_state_only",
-  "summarizedRecords": 18325,
+  "summarizedRecords": 18326,
   "version": 1
 }

manifest.json

@@ -2,24 +2,24 @@
   "agentVersion": "2026.09.22.1",
   "checksums": {
     ".modelhub_state/account_capacity.json": "930f9869793c3d1aa071c53f05b7cf82a4e372f002be88361d6d6189e27c12a1",
-    ".modelhub_state/architecture_compatibility_blacklist.json": "8a9eb6d4e19eadc5821cd7d7f4c8369965bfbaeb467025d1eeed610e547090c2",
+    ".modelhub_state/architecture_compatibility_blacklist.json": "0769a008ab47f56ee6fbe6b8acc43895aebbf71dba89e5c27d579847a59a5627",
     ".modelhub_state/architecture_history_backfill.json": "a92348206605da04f75c9e26e7c818b76f259e0b28983f18b6375ef94802f0ab",
-    ".modelhub_state/market_intelligence.json": "17182a6b6581022d95108cfdaed15d9f4059e17eac6d91aa5c38fd67fdee226a",
+    ".modelhub_state/market_intelligence.json": "3946af3abfa509f970585fc51ec556660d274f54d400a3081b2c2c7c419f3180",
-    ".modelhub_state/official_capabilities.json": "fd0e0a47bf113cadb38f398c13f734fa2d22cec485abfc556393a1c23673bf66",
+    ".modelhub_state/official_capabilities.json": "a30d12a42e9d64a5786afe3393b662f2a5c7d6c44f1572fa15bee6e7478bbecf",
-    ".modelhub_state/outcome_checkpoint.json": "564db88dbb800a8701151cddb62d1094074536d3f3131a30e00d10a44fd97962",
+    ".modelhub_state/outcome_checkpoint.json": "80b6e234b7fa51f2d395063a4f043016f7dcc103b25e2f6cbcfc9ceaba1f665c",
     ".modelhub_state/queue_cleanup_latest.json": "b1cbfbf2c5f4580663e171ef807971073749da732dfcd8a09b81c745460ba3a3",
     ".modelhub_state/recent_outcomes.jsonl": "f4376e5baf7f31d74cbefe71474a6248dca3a1ff7d0e7190fb89a708d2acd37c",
-    ".modelhub_state/recovery_active_tasks.jsonl": "2308b1b2f4875b5f8fe52d111c47c05ad75e250ec62d8b8f7d795b8fd9fb6a52",
+    ".modelhub_state/recovery_active_tasks.jsonl": "4597132c449fe78429d67b9731f6359ad9608a02510bdc210bb7194f0fd1a358",
-    ".modelhub_state/recovery_intents.jsonl": "c0c889fc0e535cb2962eef48dfc6414d72a05ed106ecd2582b9e479c9075eac1",
+    ".modelhub_state/recovery_intents.jsonl": "9bcd4ad654c7e47a5d6c96887d4e36883fc99fb110822336955ee86b5e9691c9",
     ".modelhub_state/routing_intelligence.json": "c15780438114d5fe59a4fb92bc00c3c1a5691a8c85659a742d53566a03b74474",
     ".modelhub_state/submission_exclusions.jsonl": "660ad77b924d3b0cc76f46aacd1f91a924d13a02bc68057eb5a964647441283b",
     ".modelhub_state/worker_crashes.jsonl": "b321ea3648ae6b322601ed30ef8c3b1da0f13612af1bde3a64b0d41dc18a7b9f",
     "ledger/submissions.jsonl": "c08cdb80843cf0c36585dd43563daaf9a884421d4601582f9a6016ccf4964822",
-    "outcomes/submissions.jsonl": "3e5105517f48cd85d31cb795a4e200170d172bf407a716ace7db451271a48669"
+    "outcomes/submissions.jsonl": "21e9632601fd23c54f5b260a704e765b474197a362e7cf2425413538e13ba79f"
   },
-  "generation": 23022,
+  "generation": 23023,
   "phase": "cycle",
   "schemaVersion": 1,
-  "updatedAt": "2026-10-10T14:52:22.283557+00:00",
+  "updatedAt": "2026-10-10T14:55:31.465387+00:00",
   "writerId": "43462b8a22de474db5c83d43b475a356"
 }

outcomes/submissions.jsonl

@@ -196,7 +196,6 @@
 {"failReason": null, "framework": "vllm_tokenizer_patch", "lastSyncTime": "2026-09-22T02:22:55.559751+00:00", "modelId": "nm-testing/tinyllama-oneshot-w8-channel-a8-tensor", "modelProfile": {"architectures": ["LlamaForCausalLM"], "configFingerprint": "167e80c9a74f5acd91819ac9a661546c8992cc5beb0bdd0a1dc9fc9659a5e778", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 1232059012, "estimatedRequiredGiB": 1.379, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "llama", "modelscopeFileSize": 1233907698, "modelscopeLicense": null, "modelscopeParams": 1100048384, "modelscopeTags": ["model_type:llama", "library:safetensors", "library:pytorch", "task:text-generation"], "modelscopeTasks": ["text-generation"], "quantizationMethod": "compressed-tensors", "recentProfileFeedback": {"attributableFailureCount": 0, "consecutiveFailures": 0, "decisionFailureRate": 0.0, "decisionSuccessRate": 0.0, "decisionTotal": 0, "failureBreakdown": {"ambiguous_runtime": 1}, "failureCount": 1, "failureRate": 1.0, "framework": "vllm_tokenizer_patch", "lastTerminalAt": "2026-09-21T15:02:24.876390+00:00", "modelType": "llama", "pendingCount": 0, "pendingRate": 0.0, "platformFailureCount": 0, "quantizationMethod": "compressed-tensors", "successCount": 0, "successRate": 0.0, "targetGpu": "Ascend_910-b4", "taskType": "text-generation", "total": 1, "unresolvedFailureCount": 1}, "repositoryOnDiskBytes": 1233907698}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-21T18:22:48.751608+00:00", "targetGpu": "Ascend_910-b4", "taskId": "5005404", "taskType": "text-generation", "verifyResult": null}
 {"failReason": null, "framework": "vllm_fix_tokenizer", "lastSyncTime": "2026-09-22T02:22:55.559801+00:00", "modelId": "nm-testing/llama7b-one-shot-2_4-w4a16-marlin24-t", "modelProfile": {"architectures": ["LlamaForCausalLM"], "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 3054596816, "estimatedRequiredGiB": 3.416, "gpuMemoryEvidence": {"memoryGiB": 100.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 100.0, "maximumRepositorySizeGiB": 83.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "llama", "modelscopeFileSize": 3056444703, "modelscopeLicense": null, "modelscopeParams": 1122504704, "modelscopeTags": ["model_type:llama", "library:safetensors", "library:pytorch", "task:text-generation"], "modelscopeTasks": ["text-generation"], "quantizationMethod": "compressed-tensors", "repositoryOnDiskBytes": 3056444703}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-21T18:22:48.756508+00:00", "targetGpu": "Kunlunxin_p-800", "taskId": "5005406", "taskType": "text-generation", "verifyResult": null}
 {"failReason": null, "framework": "vllm_fix_tokenizer", "lastSyncTime": "2026-09-22T02:26:33.565841+00:00", "modelId": "nm-testing/tinyllama-oneshot-w4a16-channel-v2", "modelProfile": {"architectures": ["LlamaForCausalLM"], "configFingerprint": "d8df464812ff2fd3c46dc89b0fc6a96370e694fab8e52179f1ada2cdd0ea92b9", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 747618112, "estimatedRequiredGiB": 0.838, "gpuMemoryEvidence": {"memoryGiB": 100.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 100.0, "maximumRepositorySizeGiB": 83.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "llama", "modelscopeFileSize": 749466433, "modelscopeLicense": null, "modelscopeParams": 1100048384, "modelscopeTags": ["model_type:llama", "library:safetensors", "library:pytorch", "task:text-generation"], "modelscopeTasks": ["text-generation"], "quantizationMethod": "compressed-tensors", "repositoryOnDiskBytes": 749466433}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-21T18:22:48.845075+00:00", "targetGpu": "Kunlunxin_p-800", "taskId": "5005433", "taskType": "text-generation", "verifyResult": null}
-{"failReason": null, "framework": "vllm_fix_tokenizer", "lastSyncTime": "2026-09-22T03:05:58.563225+00:00", "modelId": "nkkbr/Mini-K3-1H-decay-g4-v2_B", "modelProfile": {"architectures": [], "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 2030769952, "estimatedRequiredGiB": 2.279, "gpuMemoryEvidence": {"memoryGiB": 100.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 100.0, "maximumRepositorySizeGiB": 83.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "mini_k3", "modelscopeFileSize": null, "modelscopeLicense": null, "modelscopeParams": null, "modelscopeTags": ["model_type:mini_k3", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:kimi-k3", "custom_tag:pretraining", "custom_tag:mixture-of-experts", "custom_tag:linear-attention", "custom_tag:architecture-ablation"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 2039132310}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-21T19:05:50.451970+00:00", "targetGpu": "Kunlunxin_p-800", "taskId": "5006042", "taskType": "text-generation", "verifyResult": null}
 {"failReason": null, "framework": "vllm_fix_tokenizer", "lastSyncTime": "2026-09-22T03:05:58.563212+00:00", "modelId": "nkkbr/Mini-K3-1H-mamba2-n64-g4-v1_B", "modelProfile": {"architectures": [], "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 2009388256, "estimatedRequiredGiB": 2.25, "gpuMemoryEvidence": {"memoryGiB": 100.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 100.0, "maximumRepositorySizeGiB": 83.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "mini_k3", "modelscopeFileSize": null, "modelscopeLicense": null, "modelscopeParams": null, "modelscopeTags": ["model_type:mini_k3", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:kimi-k3", "custom_tag:mamba2", "custom_tag:pretraining", "custom_tag:mixture-of-experts", "custom_tag:architecture-ablation"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 2012995244}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-21T19:05:51.057293+00:00", "targetGpu": "Kunlunxin_p-800", "taskId": "5006049", "taskType": "text-generation", "verifyResult": null}
 {"failReason": null, "framework": "vllm", "lastSyncTime": "2026-09-22T07:42:35.356599+00:00", "modelId": "prithivMLmods/Nexus-9B-CodeCore-Merge", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "1557c77dfe4d3dc1a68ae6b8349e7d094da7eec7752d29442a9d6d27e5f13121", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 19306311000, "estimatedRequiredGiB": 21.602, "gpuMemoryEvidence": {"memoryGiB": 64.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 64.0, "maximumRepositorySizeGiB": 53.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": null, "modelscopeLicense": "apache-2.0", "modelscopeParams": 1562291952, "modelscopeTags": ["license:apache-2.0", "model_type:qwen3_5", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:text-generation-inference", "custom_tag:pytorch", "custom_tag:omnimergekit", "custom_tag:merge", "custom_tag:qwen3_5", "custom_tag:reasoning", "custom_tag:chain-of-thought", "custom_tag:sft", "custom_tag:agent", "custom_tag:tool-use", "custom_tag:function-calling", "custom_tag:coder", "custom_tag:9B"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 19329296879}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-21T23:30:46.846078+00:00", "targetGpu": "Sunrise_pt-200-x1", "taskId": "5008929", "taskType": "text-generation", "verifyResult": null}
 {"failReason": null, "framework": "vllm", "lastSyncTime": "2026-09-22T07:42:35.356417+00:00", "modelId": "prithivMLmods/Qwen-Image-2.1-PE-T2I-MLX", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "1557c77dfe4d3dc1a68ae6b8349e7d094da7eec7752d29442a9d6d27e5f13121", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 35196531592, "estimatedRequiredGiB": 39.403, "gpuMemoryEvidence": {"memoryGiB": 64.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 64.0, "maximumRepositorySizeGiB": 53.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": null, "modelscopeLicense": "other", "modelscopeParams": 1411666624, "modelscopeTags": ["license:other", "model_type:qwen3_5", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:mlx", "custom_tag:text-generation-inference", "custom_tag:qwen3_5", "custom_tag:mlx-vlm", "custom_tag:prompt-rewriter", "custom_tag:prompt enhancer", "custom_tag:apple-silicon", "custom_tag:4-bit", "custom_tag:8-bit"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 35256868927}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-21T23:32:54.642933+00:00", "targetGpu": "Sunrise_pt-200-x1", "taskId": "5008957", "taskType": "text-generation", "verifyResult": null}