state: generation 8221 (cycle)

.modelhub_state/architecture_compatibility_blacklist.json

@@ -1693,7 +1693,7 @@
       "taskType": "text-generation"
     }
   },
-  "generatedAt": "2026-09-16T14:02:36.998356+00:00",
+  "generatedAt": "2026-09-16T14:05:45.193990+00:00",
   "summary": {
     "activeBlockCount": 82,
     "byGpuFramework": {

.modelhub_state/architecture_history_backfill.json

@@ -51,7 +51,7 @@
     "4": {
       "complete": false,
       "lastError": "ModelHubAPIError: 系统错误",
-      "listingErrors": 600,
+      "listingErrors": 601,
       "nextPage": 1,
       "recordsScanned": 0,
       "uniqueRecords": 0
@@ -102,12 +102,12 @@
   "cutoffAt": "2026-09-04T03:55:51.365685+00:00",
   "failureLogsInspected": 0,
   "mode": "incremental_decision_only",
-  "nextAccountIndex": 4,
+  "nextAccountIndex": 5,
   "recordsScanned": 0,
   "seenTaskIds": [],
   "startedAt": "2026-09-04T03:55:51.365685+00:00",
   "terminalRecords": 0,
   "uniqueRecords": 0,
-  "updatedAt": "2026-09-16T14:02:36.965654+00:00",
+  "updatedAt": "2026-09-16T14:05:45.168774+00:00",
   "version": 1
 }

.modelhub_state/market_intelligence.json

@@ -416,7 +416,7 @@
     }
   },
   "frameworkUpdatedAt": null,
-  "generatedAt": "2026-09-16T14:00:55.123556+00:00",
+  "generatedAt": "2026-09-16T14:03:49.110808+00:00",
   "gpuStats": {
     "Ascend_910-b3": {
       "available": true,

.modelhub_state/outcome_checkpoint.json

@@ -1,6 +1,6 @@
 {
-  "generatedAt": "2026-09-16T13:55:12.893198+00:00",
+  "generatedAt": "2026-09-16T14:03:39.987499+00:00",
-  "lastSyncTime": "2026-09-16T13:55:11.492896+00:00",
+  "lastSyncTime": "2026-09-16T14:03:38.007248+00:00",
   "recentLimit": 300,
   "report": {
     "architectureCompatibilityBlocks": {
@@ -1762,15 +1762,15 @@
         "unresolvedFailureCount": 1
       },
       "Ascend_910-b3|vllm_tokenizer_patch|text-generation": {
-        "attributableFailureCount": 4,
+        "attributableFailureCount": 5,
         "decisionFailureRate": 1.0,
         "decisionSuccessRate": 0.0,
-        "decisionTotal": 4,
+        "decisionTotal": 5,
         "failureBreakdown": {
           "ambiguous_runtime": 4,
-          "framework_architecture_unsupported": 4
+          "framework_architecture_unsupported": 5
         },
-        "failureCount": 8,
+        "failureCount": 9,
         "failureRate": 1.0,
         "framework": "vllm_tokenizer_patch",
         "pendingCount": 0,
@@ -1780,7 +1780,7 @@
         "successRate": 0.0,
         "targetGpu": "Ascend_910-b3",
         "taskType": "text-generation",
-        "total": 8,
+        "total": 9,
         "unresolvedFailureCount": 4
       },
       "Ascend_910-b3|vllm|text-generation": {
@@ -2899,48 +2899,48 @@
         "unresolvedFailureCount": 61
       },
       "vllm_tokenizer_patch": {
-        "attributableFailureCount": 5,
+        "attributableFailureCount": 6,
         "decisionFailureRate": 1.0,
         "decisionSuccessRate": 0.0,
-        "decisionTotal": 5,
+        "decisionTotal": 6,
         "failureBreakdown": {
           "ambiguous_runtime": 5,
-          "framework_architecture_unsupported": 5
+          "framework_architecture_unsupported": 6
         },
-        "failureCount": 10,
+        "failureCount": 11,
         "failureRate": 1.0,
         "pendingCount": 0,
         "pendingRate": 0.0,
         "platformFailureCount": 0,
         "successCount": 0,
         "successRate": 0.0,
-        "total": 10,
+        "total": 11,
         "unresolvedFailureCount": 5
       }
     },
-    "generatedAt": "2026-09-16T13:55:12.886131+00:00",
+    "generatedAt": "2026-09-16T14:03:39.980991+00:00",
     "gpuSummaries": {
       "Ascend_910-b3": {
-        "attributableFailureCount": 48,
+        "attributableFailureCount": 49,
-        "decisionFailureRate": 0.96,
+        "decisionFailureRate": 0.9608,
-        "decisionSuccessRate": 0.04,
+        "decisionSuccessRate": 0.0392,
-        "decisionTotal": 50,
+        "decisionTotal": 51,
         "failureBreakdown": {
           "ambiguous_runtime": 35,
-          "framework_architecture_unsupported": 46,
+          "framework_architecture_unsupported": 47,
           "memory_capacity": 1,
           "tokenizer_compatibility": 1,
           "参数/模板问题": 9,
           "验证失败": 27
         },
-        "failureCount": 119,
+        "failureCount": 120,
-        "failureRate": 0.9835,
+        "failureRate": 0.9836,
         "pendingCount": 0,
         "pendingRate": 0.0,
         "platformFailureCount": 0,
         "successCount": 2,
-        "successRate": 0.0165,
+        "successRate": 0.0164,
-        "total": 121,
+        "total": 122,
         "unresolvedFailureCount": 71
       },
       "Ascend_910-b4": {
@@ -3341,6 +3341,29 @@
         "total": 1,
         "unresolvedFailureCount": 1
       },
+      "Ascend_910-b3|vllm_tokenizer_patch|text-generation|qwen3_5|exl3": {
+        "attributableFailureCount": 1,
+        "decisionFailureRate": 1.0,
+        "decisionSuccessRate": 0.0,
+        "decisionTotal": 1,
+        "failureBreakdown": {
+          "framework_architecture_unsupported": 1
+        },
+        "failureCount": 1,
+        "failureRate": 1.0,
+        "framework": "vllm_tokenizer_patch",
+        "modelType": "qwen3_5",
+        "pendingCount": 0,
+        "pendingRate": 0.0,
+        "platformFailureCount": 0,
+        "quantizationMethod": "exl3",
+        "successCount": 0,
+        "successRate": 0.0,
+        "targetGpu": "Ascend_910-b3",
+        "taskType": "text-generation",
+        "total": 1,
+        "unresolvedFailureCount": 0
+      },
       "Ascend_910-b4|vllm_tokenizer_patch|text-generation|lfm2|none": {
         "attributableFailureCount": 0,
         "decisionFailureRate": 0.0,
@@ -6880,6 +6903,30 @@
         "total": 1,
         "unresolvedFailureCount": 1
       },
+      "Ascend_910-b3|vllm_tokenizer_patch|text-generation|qwen3_5|exl3|33": {
+        "attributableFailureCount": 1,
+        "decisionFailureRate": 1.0,
+        "decisionSuccessRate": 0.0,
+        "decisionTotal": 1,
+        "failureBreakdown": {
+          "framework_architecture_unsupported": 1
+        },
+        "failureCount": 1,
+        "failureRate": 1.0,
+        "framework": "vllm_tokenizer_patch",
+        "loadSizeLog2Bucket": 33,
+        "modelType": "qwen3_5",
+        "pendingCount": 0,
+        "pendingRate": 0.0,
+        "platformFailureCount": 0,
+        "quantizationMethod": "exl3",
+        "successCount": 0,
+        "successRate": 0.0,
+        "targetGpu": "Ascend_910-b3",
+        "taskType": "text-generation",
+        "total": 1,
+        "unresolvedFailureCount": 0
+      },
       "Ascend_910-b4|vllm_tokenizer_patch|text-generation|lfm2|none|29": {
         "attributableFailureCount": 0,
         "decisionFailureRate": 0.0,
@@ -10849,17 +10896,17 @@
         "unresolvedFailureCount": 0
       }
     },
-    "terminalRecords": 2017,
+    "terminalRecords": 2018,
-    "totalRecords": 2111,
+    "totalRecords": 2112,
     "totals": {
-      "attributableFailureCount": 647,
+      "attributableFailureCount": 648,
-      "decisionFailureRate": 0.8912,
+      "decisionFailureRate": 0.8913,
-      "decisionSuccessRate": 0.1088,
+      "decisionSuccessRate": 0.1087,
-      "decisionTotal": 726,
+      "decisionTotal": 727,
       "failureBreakdown": {
         "ambiguous_runtime": 491,
         "backend_operator": 45,
-        "framework_architecture_unsupported": 423,
+        "framework_architecture_unsupported": 424,
         "memory_capacity": 12,
         "model_load": 84,
         "platform_infrastructure": 10,
@@ -10869,14 +10916,14 @@
         "参数/模板问题": 117,
         "验证失败": 673
       },
-      "failureCount": 1938,
+      "failureCount": 1939,
-      "failureRate": 0.9608,
+      "failureRate": 0.9609,
       "pendingCount": 0,
       "pendingRate": 0.0,
       "platformFailureCount": 10,
       "successCount": 79,
-      "successRate": 0.0392,
+      "successRate": 0.0391,
-      "total": 2017,
+      "total": 2018,
       "unresolvedFailureCount": 1281
     },
     "warnings": [
@@ -10884,13 +10931,13 @@
       "GPU Iluvatar_mrv-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-b4 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 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 hygon_k100-ai 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 Ascend_910-b3 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 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 Biren_166m|unknown|text-generation 近期失败集中，建议降低该 GPU+框架的提交优先级。",
       "组合 Biren_166m|vllm|text-generation 近期失败集中，建议降低该 GPU+框架的提交优先级。",
@@ -10916,6 +10963,6 @@
     ]
   },
   "storageMode": "decision_state_only",
-  "summarizedRecords": 2111,
+  "summarizedRecords": 2112,
   "version": 1
 }

manifest.json

@@ -1,24 +1,24 @@
 {
   "agentVersion": "2026.09.04.4",
   "checksums": {
-    ".modelhub_state/architecture_compatibility_blacklist.json": "cb24cc5c127b45f1c6d4097fbed3293d1c4cfe3bdd02019156cb8c37c19d834f",
+    ".modelhub_state/architecture_compatibility_blacklist.json": "0433383e87acbe0e480f0d9ac40431b119a54028e1b21450cce7797a03dec38d",
-    ".modelhub_state/architecture_history_backfill.json": "afdc3a9b6f592ade063379ab4764048661dd82ebd0c41f2523d1b53051261dbf",
+    ".modelhub_state/architecture_history_backfill.json": "22d7e86ef13713c2b5a573e9f365720fd407461dfefb8d608bd7ba27628668fc",
-    ".modelhub_state/market_intelligence.json": "9c78dedc5d106bb780fa940b5b4212f1f84b61e445e73cd7b23a442e30686ed5",
+    ".modelhub_state/market_intelligence.json": "160e6f2ce259cb87569c374c3bcc8a5782595789776eeaf9f5ede373f6866b07",
-    ".modelhub_state/official_capabilities.json": "d4d4e7fa00e2f4ca3febad8b34e523a4bd8b176dccb38e21de0df5978f0f0886",
+    ".modelhub_state/official_capabilities.json": "717434ad4acc8d7e3a77db3c583aff932f8cf7a5fff470fad7871d8ef67d3c55",
-    ".modelhub_state/outcome_checkpoint.json": "914b330c2974e421a9261390d1c194881b817cda1c6f55470c65e1616b381fc7",
+    ".modelhub_state/outcome_checkpoint.json": "91edade3ed851e6aeb768d6104b4b711ff5c4cb8c7845e49515c325fd377f9bb",
     ".modelhub_state/queue_cleanup_latest.json": "057079282bc277b3a97c1805915c45464c12312f8ee3ba54d0366d05f34f4466",
     ".modelhub_state/recent_outcomes.jsonl": "1c47953c8848f46a5d7f000dd779708a08016689c75a99bd80ecff0d8fcc669b",
-    ".modelhub_state/recovery_active_tasks.jsonl": "40fd65021c8dac7bba428a24bb39c83592728f5b692c5e6588268c4ca4f4fe76",
+    ".modelhub_state/recovery_active_tasks.jsonl": "e26cb3b940df04e297c3ffbc4ddfa643de273fd5d087a5303a6e467a1f9e3457",
-    ".modelhub_state/recovery_intents.jsonl": "040ba88a392d97fd1f201baea25da79061538d7d9715bea6c8c3c2b313e41204",
+    ".modelhub_state/recovery_intents.jsonl": "88e9d87e194a325c35a997db4dcf5f3272aadb5671c6ee821fb1b793df5c9f77",
     ".modelhub_state/routing_intelligence.json": "6eb0869a3775dab84161e373d7a23bd67d2843b691074760b256dca93142d124",
     ".modelhub_state/submission_exclusions.jsonl": "632739c6cd6a2dd2237572ba18e6a390c5aa37ce56a18a2c1341d78569421552",
     ".modelhub_state/worker_crashes.jsonl": "b41d1dda7bab11bedd96de40fecd2d416e47396901b3283f48a56de5d6348983",
     "ledger/submissions.jsonl": "1a49861c72001064b378e98b55269416e82d59164cbdff44740b3a41cd51b00e",
-    "outcomes/submissions.jsonl": "a44e219eb53e86e45923ad63a713088668dd2228bb61708258c32208a0cefc6f"
+    "outcomes/submissions.jsonl": "ca14c26ce5c7d7b172905c62fb42591fad0f19eb29c4b63f8333d9c1cf96ad86"
   },
-  "generation": 8220,
+  "generation": 8221,
   "phase": "cycle",
   "schemaVersion": 1,
-  "updatedAt": "2026-09-16T14:02:37.114916+00:00",
+  "updatedAt": "2026-09-16T14:05:45.733605+00:00",
   "writerId": "eccb3e0018f640d19e578c271a207b5c"
 }

outcomes/submissions.jsonl

@@ -10,7 +10,6 @@
 {"failReason": null, "framework": "vllm", "lastSyncTime": "2026-09-04T14:58:13.792120+00:00", "modelId": "Jackrong/DeepSeek-V4-Pro-Qwen3.5-9B", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "d341b8002cf39d3299fd429479e8c8e954583e49d4fec5a640ddc0f8f0cbd054", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 19306310880, "estimatedRequiredGiB": 21.599, "gpuMemoryEvidence": {"memoryGiB": 64.0, "source": "manufacturer_spec", "sourceUrl": "https://www.birentech.com/news/id6rz98v3obczy77cmxzfgk3/"}, "gpuMemoryGiB": 64.0, "maximumRepositorySizeGiB": 53.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": 19326449341, "modelscopeLicense": "apache-2.0", "modelscopeParams": 9653104368, "modelscopeTags": ["license:apache-2.0", "model_type:qwen3_5", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:text-generation-inference", "custom_tag:transformers", "custom_tag:unsloth", "custom_tag:qwen3_5", "custom_tag:reasoning", "custom_tag:distillation", "custom_tag:deepseek", "custom_tag:sft", "custom_tag:rl", "custom_tag:gspo", "custom_tag:math", "custom_tag:stem", "custom_tag:tool-use", "custom_tag:function-calling", "custom_tag:mtp"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 19326449341}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-04T06:55:47.742909+00:00", "targetGpu": "Biren_166m", "taskId": "4621594", "taskType": "text-generation", "verifyResult": null}
 {"failReason": null, "framework": "vllm", "lastSyncTime": "2026-09-04T14:58:13.792154+00:00", "modelId": "VerStella/Qwen3.8-27B-heretic-ara-W8A8-DCU", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "d341b8002cf39d3299fd429479e8c8e954583e49d4fec5a640ddc0f8f0cbd054", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 31228126952, "estimatedRequiredGiB": 34.934, "gpuMemoryEvidence": {"memoryGiB": 64.0, "source": "manufacturer_spec", "sourceUrl": "https://www.birentech.com/news/id6rz98v3obczy77cmxzfgk3/"}, "gpuMemoryGiB": 64.0, "maximumRepositorySizeGiB": 53.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": 31258377392, "modelscopeLicense": "apache-2.0", "modelscopeParams": 27781427952, "modelscopeTags": ["license:apache-2.0", "model_type:qwen3_5", "library:transformer", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:qwen", "custom_tag:qwen3.8", "custom_tag:w8a8", "custom_tag:int8", "custom_tag:quantized"], "modelscopeTasks": ["text-generation"], "quantizationMethod": "compressed-tensors", "repositoryOnDiskBytes": 31258377392}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-04T06:55:47.759663+00:00", "targetGpu": "Biren_166m", "taskId": "4621595", "taskType": "text-generation", "verifyResult": null}
 {"failReason": null, "framework": "vllm", "lastSyncTime": "2026-09-04T15:14:29.722109+00:00", "modelId": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Quality-FP16", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "d341b8002cf39d3299fd429479e8c8e954583e49d4fec5a640ddc0f8f0cbd054", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 29952487299, "estimatedRequiredGiB": 33.498, "gpuMemoryEvidence": {"memoryGiB": 64.0, "source": "manufacturer_spec", "sourceUrl": "https://www.birentech.com/news/id6rz98v3obczy77cmxzfgk3/"}, "gpuMemoryGiB": 64.0, "maximumRepositorySizeGiB": 53.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": 29973198172, "modelscopeLicense": "apache-2.0", "modelscopeParams": 8027131120, "modelscopeTags": ["license:apache-2.0", "model_type:qwen3_5", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:mlx", "custom_tag:apple-silicon", "custom_tag:macos", "custom_tag:m1", "custom_tag:m2", "custom_tag:fp16", "custom_tag:speculative-decoding", "custom_tag:multi-token-prediction", "custom_tag:qwen", "custom_tag:qwen3.8", "custom_tag:mtp", "custom_tag:mtplx", "custom_tag:local-ai", "custom_tag:coding"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 29973198172}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-04T07:12:30.054811+00:00", "targetGpu": "Biren_166m", "taskId": "4621809", "taskType": "text-generation", "verifyResult": null}
-{"failReason": null, "framework": "vllm_tokenizer_patch", "lastSyncTime": "2026-09-04T15:58:32.287733+00:00", "modelId": "Mia-AiLab/Qwen3.8-27B-EXL3-3.5bpw", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "7b5500238d52e9e2a21471d722bc1636ca047e1a12ba49c491593b86bec602b8", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 15338408461, "estimatedRequiredGiB": 17.168, "gpuMemoryEvidence": {"memoryGiB": 64.0, "source": "published_card_spec", "sourceUrl": "https://aclanthology.org/2025.emnlp-main.1630.pdf"}, "gpuMemoryGiB": 64.0, "maximumRepositorySizeGiB": 53.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": 15362064483, "modelscopeLicense": "apache-2.0", "modelscopeParams": 7669052656, "modelscopeTags": ["license:apache-2.0", "model_type:qwen3_5", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:exl3", "custom_tag:exllamav3", "custom_tag:quantization", "custom_tag:long-context", "custom_tag:speculative-decoding"], "modelscopeTasks": ["text-generation"], "quantizationMethod": "exl3", "repositoryOnDiskBytes": 15362064483}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-04T07:56:50.952206+00:00", "targetGpu": "Ascend_910-b3", "taskId": "4622417", "taskType": "text-generation", "verifyResult": null}
 {"failReason": null, "framework": "vllm-mlu", "lastSyncTime": "2026-09-04T16:14:29.511220+00:00", "modelId": "Mia-AiLab/Qwen3.8-27B-EXL3-3.5bpw", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "36ec94e289a8da5fd70b98a2f401fc5159740daf9e2a076da4fd490f34bcbc0b", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 15338408461, "estimatedRequiredGiB": 17.168, "gpuMemoryEvidence": {"memoryGiB": 24.0, "source": "modelhub_preflight_oom:49"}, "gpuMemoryGiB": 24.0, "maximumRepositorySizeGiB": 20.0, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": 15362064483, "modelscopeLicense": "apache-2.0", "modelscopeParams": 7669052656, "modelscopeTags": ["license:apache-2.0", "model_type:qwen3_5", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:exl3", "custom_tag:exllamav3", "custom_tag:quantization", "custom_tag:long-context", "custom_tag:speculative-decoding"], "modelscopeTasks": ["text-generation"], "quantizationMethod": "exl3", "repositoryOnDiskBytes": 15362064483}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-04T08:14:12.544960+00:00", "targetGpu": "Cambricon_mlu-370-x4", "taskId": "4622655", "taskType": "text-generation", "verifyResult": null}
 {"failReason": null, "framework": "vllm", "lastSyncTime": "2026-09-04T16:33:41.731498+00:00", "modelId": "Mia-AiLab/Qwen3.8-27B-EXL3-3.5bpw", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "d341b8002cf39d3299fd429479e8c8e954583e49d4fec5a640ddc0f8f0cbd054", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 15338408461, "estimatedRequiredGiB": 17.168, "gpuMemoryEvidence": {"memoryGiB": 64.0, "source": "manufacturer_spec", "sourceUrl": "https://www.birentech.com/news/id6rz98v3obczy77cmxzfgk3/"}, "gpuMemoryGiB": 64.0, "maximumRepositorySizeGiB": 53.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": 15362064483, "modelscopeLicense": "apache-2.0", "modelscopeParams": 7669052656, "modelscopeTags": ["license:apache-2.0", "model_type:qwen3_5", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:exl3", "custom_tag:exllamav3", "custom_tag:quantization", "custom_tag:long-context", "custom_tag:speculative-decoding"], "modelscopeTasks": ["text-generation"], "quantizationMethod": "exl3", "repositoryOnDiskBytes": 15362064483}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-04T08:33:06.846794+00:00", "targetGpu": "Biren_166m", "taskId": "4623002", "taskType": "text-generation", "verifyResult": null}
 {"failReason": null, "framework": "vllm_tokenizer_patch", "lastSyncTime": "2026-09-04T20:39:27.718139+00:00", "modelId": "ornith-ai/Ornith-1.5-9B-NVFP4", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "167e80c9a74f5acd91819ac9a661546c8992cc5beb0bdd0a1dc9fc9659a5e778", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 8818103892, "estimatedRequiredGiB": 9.883, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "modelhub_preflight_oom:8"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": 8842796317, "modelscopeLicense": "mit", "modelscopeParams": 6728625904, "modelscopeTags": ["license:mit", "model_type:qwen3_5", "library:safetensors", "library:pytorch", "task:text-generation"], "modelscopeTasks": ["text-generation"], "quantizationMethod": "modelopt", "repositoryOnDiskBytes": 8842796317}, "outcome": "pending", "status": "waiting", "submitTime": "2026-09-04T12:34:49.403585+00:00", "targetGpu": "Ascend_910-b4", "taskId": "4626360", "taskType": "text-generation", "verifyResult": null}