diff --git a/.modelhub_state/recovery_intents.jsonl b/.modelhub_state/recovery_intents.jsonl
index 2d96a47ac1..7a6e35aed4 100644
--- a/.modelhub_state/recovery_intents.jsonl
+++ b/.modelhub_state/recovery_intents.jsonl
@@ -1534,24 +1534,24 @@
 {"batchId": "40023a20f9cf4cf8bd1d244e1cc039ec", "completedAt": "2026-09-20T04:46:08.050460+00:00", "configFingerprint": "1cc16a7df03cf7b1485e2fda7e5c05e286a546d04e80bdd595c015004b3852a8", "configSource": "modelhub_live", "createdAt": "2026-09-20T04:46:06.830336+00:00", "framework": "vllm_0_17_0_corex_4_4_0", "intentId": "d43a0c03344441919e2024b8a8b63c3f", "lastModified": "2026-09-16T16:38:45+00:00", "modelAddress": "https://modelscope.cn/models/RedHatAI/gemma-4-12B-it-FP8-Dynamic", "reason": null, "reconciledAt": "2026-09-20T05:15:18.531852+00:00", "repoId": "RedHatAI/gemma-4-12B-it-FP8-Dynamic", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "recovered_active", "targetGpu": "Iluvatar_bi-150", "taskId": "4970952", "taskType": "text-generation"}
 {"batchId": "9cd422eaba2d4d5d986041ec575cafd4", "completedAt": "2026-09-20T05:00:40.986239+00:00", "configFingerprint": "1cc16a7df03cf7b1485e2fda7e5c05e286a546d04e80bdd595c015004b3852a8", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:00:40.023437+00:00", "framework": "vllm_0_17_0_corex_4_4_0", "intentId": "df9343932b5840e5a9eaa987bf2c6f17", "lastModified": "2026-09-16T17:07:54+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/LFM2.5-1.2B-Thinking-6bit", "reason": null, "reconciledAt": "2026-09-20T05:15:18.531662+00:00", "repoId": "mlx-community/LFM2.5-1.2B-Thinking-6bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "recovered_active", "targetGpu": "Iluvatar_bi-150", "taskId": "4971125", "taskType": "text-generation"}
 {"batchId": "bcf9a40c95a44ac29abb414ae72cc7b0", "completedAt": "2026-09-20T05:01:56.147848+00:00", "configFingerprint": "d341b8002cf39d3299fd429479e8c8e954583e49d4fec5a640ddc0f8f0cbd054", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:01:55.270355+00:00", "framework": "vllm", "intentId": "2a321630f870431cb4d1496442b9c912", "lastModified": "2026-09-16T16:38:45+00:00", "modelAddress": "https://modelscope.cn/models/RedHatAI/gemma-4-12B-it-FP8-Dynamic", "reason": null, "reconciledAt": "2026-09-20T05:15:18.533094+00:00", "repoId": "RedHatAI/gemma-4-12B-it-FP8-Dynamic", "safeConfigVector": {"dtype": "-", "gpuMemoryUtilization": 0.9, "gpuNum": 1, "maxModelLen": 4096, "tensorParallel": 1}, "status": "recovered_active", "targetGpu": "Biren_166m", "taskId": "4971137", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.266246+00:00", "framework": "vllm_fix_tokenizer", "intentId": "26d50ae46d7d42e9b1fd706121dcc344", "lastModified": "2026-09-15T14:59:20+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-12B-it-qat-OptiQ-4bit", "repoId": "mlx-community/gemma-4-12B-it-qat-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267206+00:00", "framework": "vllm_fix_tokenizer", "intentId": "db95a826d4504e65bea9d673f6abdba7", "lastModified": "2026-09-15T15:08:08+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/KAT-Coder-V2.5-Dev-OptiQ-4bit", "repoId": "mlx-community/KAT-Coder-V2.5-Dev-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267266+00:00", "framework": "vllm_fix_tokenizer", "intentId": "4d0d0f0130464d1c91e39c021d7dbe94", "lastModified": "2026-09-15T15:01:23+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-26B-A4B-it-qat-OptiQ-4bit", "repoId": "mlx-community/gemma-4-26B-A4B-it-qat-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267312+00:00", "framework": "vllm_fix_tokenizer", "intentId": "f39f433bda784940886eee83ced8e8a2", "lastModified": "2026-09-17T15:15:04+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-31B-it-qat-OptiQ-4bit", "repoId": "mlx-community/gemma-4-31B-it-qat-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267355+00:00", "framework": "vllm_fix_tokenizer", "intentId": "cf5406428bd54776af7fa2a5ff5678c0", "lastModified": "2026-09-15T14:57:59+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/Qwen3.5-122B-A10B-OptiQ-2bit-REAP-63B", "repoId": "mlx-community/Qwen3.5-122B-A10B-OptiQ-2bit-REAP-63B", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267398+00:00", "framework": "vllm_fix_tokenizer", "intentId": "0480d8a0745c40d7b18b2e10df1ecb9c", "lastModified": "2026-09-15T15:11:14+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit", "repoId": "mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267441+00:00", "framework": "transformers", "intentId": "9a4822cbd9714f65810d2c83b885cbfa", "lastModified": "2026-09-17T14:05:32+00:00", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed", "repoId": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267507+00:00", "framework": "transformers", "intentId": "886fd2e922a5487c8e2db4dec8b4d720", "lastModified": "2026-09-14T19:52:04+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-e4b-it-OptiQ-4bit", "repoId": "mlx-community/gemma-4-e4b-it-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267571+00:00", "framework": "transformers", "intentId": "0014c8782b674866894328fed6925677", "lastModified": "2026-09-17T12:51:11+00:00", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed-FP16", "repoId": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed-FP16", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267634+00:00", "framework": "transformers", "intentId": "1b8bec3a65e7414e96d981c10422f435", "lastModified": "2026-09-17T13:39:53+00:00", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed", "repoId": "Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267702+00:00", "framework": "transformers", "intentId": "ae238d8c09454e45a1f7ff05ccc87c20", "lastModified": "2026-09-14T19:46:18+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-e2b-it-OptiQ-4bit", "repoId": "mlx-community/gemma-4-e2b-it-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267763+00:00", "framework": "transformers", "intentId": "f7cdcad905e2453dbb6659f2f4a2a5a6", "lastModified": "2026-09-19T12:07:43+00:00", "modelAddress": "https://modelscope.cn/models/Arain119/Sophia", "repoId": "Arain119/Sophia", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267826+00:00", "framework": "transformers", "intentId": "72abf1e6a7174129a1d75c8b9d7c807b", "lastModified": "2026-09-17T13:54:18+00:00", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed-FP16", "repoId": "Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed-FP16", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267890+00:00", "framework": "transformers", "intentId": "468c28cce391422585b4fd25df89208c", "lastModified": "2026-09-16T17:02:01+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/Devstral-Small-2-24B-Instruct-2512-OptiQ-4bit", "repoId": "mlx-community/Devstral-Small-2-24B-Instruct-2512-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267951+00:00", "framework": "transformers", "intentId": "3776352c74d14281976c956a4fde5dd2", "lastModified": "2026-09-16T17:13:44+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/LFM2.5-1.2B-Instruct-4bit", "repoId": "mlx-community/LFM2.5-1.2B-Instruct-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.268012+00:00", "framework": "transformers", "intentId": "b821c772821a4dfa954c0e32a994be11", "lastModified": "2026-09-17T13:29:50+00:00", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.5-9B-MTPLX-Optimized-Speed", "repoId": "Youssofal/Qwen3.5-9B-MTPLX-Optimized-Speed", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "0d5bc82787a89d4313672808d1a8629a1fc675c91aa6a08865f770aa24447fbc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.268073+00:00", "framework": "vllm-patch-tokenizer", "intentId": "4319930e38ee4c9581dd8bf58ed771a6", "lastModified": "2026-09-16T16:38:45+00:00", "modelAddress": "https://modelscope.cn/models/RedHatAI/gemma-4-12B-it-FP8-Dynamic", "repoId": "RedHatAI/gemma-4-12B-it-FP8-Dynamic", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "hygon_k100-ai", "taskType": "text-generation"}
-{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "configFingerprint": "ac4b2845ddd4bdce478dd2b65134ee3ab3d259a119afe6841bf068b135fd4892", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.268116+00:00", "framework": "vllm", "intentId": "5b99cfaa8b6844a0ac731ac6b1054f5e", "lastModified": "2026-09-16T16:39:48+00:00", "modelAddress": "https://modelscope.cn/models/BAAI/RoboBrain2.5-8B-MT", "repoId": "BAAI/RoboBrain2.5-8B-MT", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "pending", "targetGpu": "Iluvatar_bi-150", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459935+00:00", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.266246+00:00", "framework": "vllm_fix_tokenizer", "intentId": "26d50ae46d7d42e9b1fd706121dcc344", "lastModified": "2026-09-15T14:59:20+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-12B-it-qat-OptiQ-4bit", "reason": null, "repoId": "mlx-community/gemma-4-12B-it-qat-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971306", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459946+00:00", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267206+00:00", "framework": "vllm_fix_tokenizer", "intentId": "db95a826d4504e65bea9d673f6abdba7", "lastModified": "2026-09-15T15:08:08+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/KAT-Coder-V2.5-Dev-OptiQ-4bit", "reason": null, "repoId": "mlx-community/KAT-Coder-V2.5-Dev-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971298", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459950+00:00", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267266+00:00", "framework": "vllm_fix_tokenizer", "intentId": "4d0d0f0130464d1c91e39c021d7dbe94", "lastModified": "2026-09-15T15:01:23+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-26B-A4B-it-qat-OptiQ-4bit", "reason": null, "repoId": "mlx-community/gemma-4-26B-A4B-it-qat-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971299", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459953+00:00", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267312+00:00", "framework": "vllm_fix_tokenizer", "intentId": "f39f433bda784940886eee83ced8e8a2", "lastModified": "2026-09-17T15:15:04+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-31B-it-qat-OptiQ-4bit", "reason": null, "repoId": "mlx-community/gemma-4-31B-it-qat-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971300", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459957+00:00", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267355+00:00", "framework": "vllm_fix_tokenizer", "intentId": "cf5406428bd54776af7fa2a5ff5678c0", "lastModified": "2026-09-15T14:57:59+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/Qwen3.5-122B-A10B-OptiQ-2bit-REAP-63B", "reason": null, "repoId": "mlx-community/Qwen3.5-122B-A10B-OptiQ-2bit-REAP-63B", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971297", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459960+00:00", "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267398+00:00", "framework": "vllm_fix_tokenizer", "intentId": "0480d8a0745c40d7b18b2e10df1ecb9c", "lastModified": "2026-09-15T15:11:14+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit", "reason": null, "repoId": "mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971308", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459963+00:00", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267441+00:00", "framework": "transformers", "intentId": "9a4822cbd9714f65810d2c83b885cbfa", "lastModified": "2026-09-17T14:05:32+00:00", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed", "reason": null, "repoId": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971303", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459968+00:00", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267507+00:00", "framework": "transformers", "intentId": "886fd2e922a5487c8e2db4dec8b4d720", "lastModified": "2026-09-14T19:52:04+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-e4b-it-OptiQ-4bit", "reason": null, "repoId": "mlx-community/gemma-4-e4b-it-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971302", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459971+00:00", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267571+00:00", "framework": "transformers", "intentId": "0014c8782b674866894328fed6925677", "lastModified": "2026-09-17T12:51:11+00:00", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed-FP16", "reason": null, "repoId": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed-FP16", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971307", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459974+00:00", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267634+00:00", "framework": "transformers", "intentId": "1b8bec3a65e7414e96d981c10422f435", "lastModified": "2026-09-17T13:39:53+00:00", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed", "reason": null, "repoId": "Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971304", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459977+00:00", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267702+00:00", "framework": "transformers", "intentId": "ae238d8c09454e45a1f7ff05ccc87c20", "lastModified": "2026-09-14T19:46:18+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-e2b-it-OptiQ-4bit", "reason": null, "repoId": "mlx-community/gemma-4-e2b-it-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971305", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459980+00:00", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267763+00:00", "framework": "transformers", "intentId": "f7cdcad905e2453dbb6659f2f4a2a5a6", "lastModified": "2026-09-19T12:07:43+00:00", "modelAddress": "https://modelscope.cn/models/Arain119/Sophia", "reason": null, "repoId": "Arain119/Sophia", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971301", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459983+00:00", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267826+00:00", "framework": "transformers", "intentId": "72abf1e6a7174129a1d75c8b9d7c807b", "lastModified": "2026-09-17T13:54:18+00:00", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed-FP16", "reason": null, "repoId": "Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed-FP16", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971309", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459985+00:00", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267890+00:00", "framework": "transformers", "intentId": "468c28cce391422585b4fd25df89208c", "lastModified": "2026-09-16T17:02:01+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/Devstral-Small-2-24B-Instruct-2512-OptiQ-4bit", "reason": null, "repoId": "mlx-community/Devstral-Small-2-24B-Instruct-2512-OptiQ-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971310", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459987+00:00", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.267951+00:00", "framework": "transformers", "intentId": "3776352c74d14281976c956a4fde5dd2", "lastModified": "2026-09-16T17:13:44+00:00", "modelAddress": "https://modelscope.cn/models/mlx-community/LFM2.5-1.2B-Instruct-4bit", "reason": null, "repoId": "mlx-community/LFM2.5-1.2B-Instruct-4bit", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971312", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459989+00:00", "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.268012+00:00", "framework": "transformers", "intentId": "b821c772821a4dfa954c0e32a994be11", "lastModified": "2026-09-17T13:29:50+00:00", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.5-9B-MTPLX-Optimized-Speed", "reason": null, "repoId": "Youssofal/Qwen3.5-9B-MTPLX-Optimized-Speed", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971314", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459992+00:00", "configFingerprint": "0d5bc82787a89d4313672808d1a8629a1fc675c91aa6a08865f770aa24447fbc", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.268073+00:00", "framework": "vllm-patch-tokenizer", "intentId": "4319930e38ee4c9581dd8bf58ed771a6", "lastModified": "2026-09-16T16:38:45+00:00", "modelAddress": "https://modelscope.cn/models/RedHatAI/gemma-4-12B-it-FP8-Dynamic", "reason": null, "repoId": "RedHatAI/gemma-4-12B-it-FP8-Dynamic", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "hygon_k100-ai", "taskId": "4971311", "taskType": "text-generation"}
+{"batchId": "b14f9b38b0c54b9f830cf71cd274ce28", "completedAt": "2026-09-20T05:16:13.459994+00:00", "configFingerprint": "ac4b2845ddd4bdce478dd2b65134ee3ab3d259a119afe6841bf068b135fd4892", "configSource": "modelhub_live", "createdAt": "2026-09-20T05:16:04.268116+00:00", "framework": "vllm", "intentId": "5b99cfaa8b6844a0ac731ac6b1054f5e", "lastModified": "2026-09-16T16:39:48+00:00", "modelAddress": "https://modelscope.cn/models/BAAI/RoboBrain2.5-8B-MT", "reason": null, "repoId": "BAAI/RoboBrain2.5-8B-MT", "safeConfigVector": {"gpuNum": 1, "maxModelLen": 4096}, "status": "submitted", "targetGpu": "Iluvatar_bi-150", "taskId": "4971313", "taskType": "text-generation"}
 {"batchId": "9b66f3a164ed48ca9a534db1d291df0f", "completedAt": "2026-09-20T04:41:55.736464+00:00", "configFingerprint": "2ebd1d3ee8ed952dd08e24120ad5d7342c9a018f55c478225a242b664721e8d0", "configSource": "modelhub_live", "createdAt": "2026-09-20T04:36:47.619879+00:00", "framework": "vllm", "intentId": "6b3a8de2b10a428c9ddf9d22b3181296", "repoId": "RedHatAI/starcoder2-15b-FP8", "safeConfigVector": {"dtype": "-", "gpuNum": 1, "maxModelLen": 2048}, "status": "age_policy_deferred", "targetGpu": "Ascend_910-b4", "taskId": null, "taskType": "text-generation"}
 {"batchId": "9b66f3a164ed48ca9a534db1d291df0f", "completedAt": "2026-09-20T04:41:55.736461+00:00", "configFingerprint": "2ebd1d3ee8ed952dd08e24120ad5d7342c9a018f55c478225a242b664721e8d0", "configSource": "modelhub_live", "createdAt": "2026-09-20T04:36:47.619821+00:00", "framework": "vllm", "intentId": "b9843c023d764a80b3d64be3bd78f7e3", "repoId": "neuralmagic/starcoder2-15b-FP8", "safeConfigVector": {"dtype": "-", "gpuNum": 1, "maxModelLen": 2048}, "status": "age_policy_deferred", "targetGpu": "Ascend_910-b4", "taskId": null, "taskType": "text-generation"}
 {"batchId": "9b66f3a164ed48ca9a534db1d291df0f", "completedAt": "2026-09-20T04:41:55.736459+00:00", "configFingerprint": "58bd440e460cf3e4d9271b3c32c451e80254e59faf47ce55dcda11588cf4117e", "configSource": "modelhub_live", "createdAt": "2026-09-20T04:36:47.619764+00:00", "framework": "vllm", "intentId": "02e8d73581314d238754c645ce8907e2", "repoId": "neuralmagic/SmolLM-1.7B-Instruct-quantized.w8a8", "safeConfigVector": {"dtype": "-", "gpuNum": 1, "maxModelLen": 2048}, "status": "age_policy_deferred", "targetGpu": "Mthreads_s4000", "taskId": null, "taskType": "text-generation"}
diff --git a/.modelhub_state/routing_intelligence.json b/.modelhub_state/routing_intelligence.json
index dfbf860581..8bf375b17a 100644
--- a/.modelhub_state/routing_intelligence.json
+++ b/.modelhub_state/routing_intelligence.json
@@ -1,6 +1,6 @@
 {
   "acceptedByCategory": {
-    "unified_success_first": 1674
+    "unified_success_first": 1692
   },
   "acceptedByRoute": {
     "text-generation|Ascend_910-b3|llamacpp": 17,
@@ -18,10 +18,10 @@
     "text-generation|Cambricon_mlu-370-x8|vllm-customized": 24,
     "text-generation|Cambricon_mlu-370-x8|vllm-mlu": 88,
     "text-generation|Iluvatar_bi-150|llamacpp": 16,
-    "text-generation|Iluvatar_bi-150|transformers": 6,
-    "text-generation|Iluvatar_bi-150|vllm": 6,
+    "text-generation|Iluvatar_bi-150|transformers": 16,
+    "text-generation|Iluvatar_bi-150|vllm": 7,
     "text-generation|Iluvatar_bi-150|vllm_0_17_0_corex_4_4_0": 101,
-    "text-generation|Iluvatar_bi-150|vllm_fix_tokenizer": 25,
+    "text-generation|Iluvatar_bi-150|vllm_fix_tokenizer": 31,
     "text-generation|Iluvatar_mrv-100|vllm": 61,
     "text-generation|Kunlunxin_p-800|vllm": 3,
     "text-generation|Kunlunxin_p-800|vllm_fix_tokenizer": 164,
@@ -34,10 +34,10 @@
     "text-generation|Vastai_va16|vllm_fix_tokenizer": 73,
     "text-generation|hygon_k100-ai|llamacpp": 18,
     "text-generation|hygon_k100-ai|vllm": 26,
-    "text-generation|hygon_k100-ai|vllm-patch-tokenizer": 106
+    "text-generation|hygon_k100-ai|vllm-patch-tokenizer": 107
   },
-  "acceptedSinceRefresh": 1674,
-  "acceptedTotal": 1674,
-  "generatedAt": "2026-09-20T05:01:56.091863+00:00",
+  "acceptedSinceRefresh": 1692,
+  "acceptedTotal": 1692,
+  "generatedAt": "2026-09-20T05:16:13.411277+00:00",
   "version": 1
 }
diff --git a/ledger/submissions.jsonl b/ledger/submissions.jsonl
index 910d16cd02..a5a8f68232 100644
--- a/ledger/submissions.jsonl
+++ b/ledger/submissions.jsonl
@@ -1033,3 +1033,21 @@
 {"framework": "vllm_fix_tokenizer", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-e2b-it-OptiQ-4bit", "modelId": "mlx-community/gemma-4-e2b-it-OptiQ-4bit", "submitTime": "2026-09-20T04:27:06.277205+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4970734", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-vllm-fix-tokenizer-iluvatar-bi-150"}
 {"framework": "vllm_fix_tokenizer", "modelAddress": "https://modelscope.cn/models/mlx-community/LFM2.5-1.2B-Instruct-4bit", "modelId": "mlx-community/LFM2.5-1.2B-Instruct-4bit", "submitTime": "2026-09-20T04:31:05.666143+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4970790", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-vllm-fix-tokenizer-iluvatar-bi-150"}
 {"framework": "vllm_fix_tokenizer", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-e4b-it-OptiQ-4bit", "modelId": "mlx-community/gemma-4-e4b-it-OptiQ-4bit", "submitTime": "2026-09-20T04:42:19.137834+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4970925", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-vllm-fix-tokenizer-iluvatar-bi-150"}
+{"framework": "vllm_fix_tokenizer", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-12B-it-qat-OptiQ-4bit", "modelId": "mlx-community/gemma-4-12B-it-qat-OptiQ-4bit", "submitTime": "2026-09-20T05:16:13.156477+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971306", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-vllm-fix-tokenizer-iluvatar-bi-150"}
+{"framework": "vllm_fix_tokenizer", "modelAddress": "https://modelscope.cn/models/mlx-community/KAT-Coder-V2.5-Dev-OptiQ-4bit", "modelId": "mlx-community/KAT-Coder-V2.5-Dev-OptiQ-4bit", "submitTime": "2026-09-20T05:16:13.138764+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971298", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-vllm-fix-tokenizer-iluvatar-bi-150"}
+{"framework": "vllm_fix_tokenizer", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-26B-A4B-it-qat-OptiQ-4bit", "modelId": "mlx-community/gemma-4-26B-A4B-it-qat-OptiQ-4bit", "submitTime": "2026-09-20T05:16:13.142829+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971299", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-vllm-fix-tokenizer-iluvatar-bi-150"}
+{"framework": "vllm_fix_tokenizer", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-31B-it-qat-OptiQ-4bit", "modelId": "mlx-community/gemma-4-31B-it-qat-OptiQ-4bit", "submitTime": "2026-09-20T05:16:13.136785+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971300", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-vllm-fix-tokenizer-iluvatar-bi-150"}
+{"framework": "vllm_fix_tokenizer", "modelAddress": "https://modelscope.cn/models/mlx-community/Qwen3.5-122B-A10B-OptiQ-2bit-REAP-63B", "modelId": "mlx-community/Qwen3.5-122B-A10B-OptiQ-2bit-REAP-63B", "submitTime": "2026-09-20T05:16:13.139470+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971297", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-vllm-fix-tokenizer-iluvatar-bi-150"}
+{"framework": "vllm_fix_tokenizer", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit", "modelId": "mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit", "submitTime": "2026-09-20T05:16:13.160907+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971308", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-vllm-fix-tokenizer-iluvatar-bi-150"}
+{"framework": "transformers", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed", "modelId": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed", "submitTime": "2026-09-20T05:16:13.144215+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971303", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-transformers-iluvatar-bi-150"}
+{"framework": "transformers", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-e4b-it-OptiQ-4bit", "modelId": "mlx-community/gemma-4-e4b-it-OptiQ-4bit", "submitTime": "2026-09-20T05:16:13.158936+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971302", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-transformers-iluvatar-bi-150"}
+{"framework": "transformers", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed-FP16", "modelId": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed-FP16", "submitTime": "2026-09-20T05:16:13.154796+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971307", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-transformers-iluvatar-bi-150"}
+{"framework": "transformers", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed", "modelId": "Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed", "submitTime": "2026-09-20T05:16:13.140867+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971304", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-transformers-iluvatar-bi-150"}
+{"framework": "transformers", "modelAddress": "https://modelscope.cn/models/mlx-community/gemma-4-e2b-it-OptiQ-4bit", "modelId": "mlx-community/gemma-4-e2b-it-OptiQ-4bit", "submitTime": "2026-09-20T05:16:13.149648+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971305", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-transformers-iluvatar-bi-150"}
+{"framework": "transformers", "modelAddress": "https://modelscope.cn/models/Arain119/Sophia", "modelId": "Arain119/Sophia", "submitTime": "2026-09-20T05:16:13.151906+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971301", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-transformers-iluvatar-bi-150"}
+{"framework": "transformers", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed-FP16", "modelId": "Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed-FP16", "submitTime": "2026-09-20T05:16:13.300793+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971309", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-transformers-iluvatar-bi-150"}
+{"framework": "transformers", "modelAddress": "https://modelscope.cn/models/mlx-community/Devstral-Small-2-24B-Instruct-2512-OptiQ-4bit", "modelId": "mlx-community/Devstral-Small-2-24B-Instruct-2512-OptiQ-4bit", "submitTime": "2026-09-20T05:16:13.323221+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971310", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-transformers-iluvatar-bi-150"}
+{"framework": "transformers", "modelAddress": "https://modelscope.cn/models/mlx-community/LFM2.5-1.2B-Instruct-4bit", "modelId": "mlx-community/LFM2.5-1.2B-Instruct-4bit", "submitTime": "2026-09-20T05:16:13.330122+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971312", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-transformers-iluvatar-bi-150"}
+{"framework": "transformers", "modelAddress": "https://modelscope.cn/models/Youssofal/Qwen3.5-9B-MTPLX-Optimized-Speed", "modelId": "Youssofal/Qwen3.5-9B-MTPLX-Optimized-Speed", "submitTime": "2026-09-20T05:16:13.358703+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971314", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-transformers-iluvatar-bi-150"}
+{"framework": "vllm-patch-tokenizer", "modelAddress": "https://modelscope.cn/models/RedHatAI/gemma-4-12B-it-FP8-Dynamic", "modelId": "RedHatAI/gemma-4-12B-it-FP8-Dynamic", "submitTime": "2026-09-20T05:16:13.324611+00:00", "targetGpu": "hygon_k100-ai", "taskId": "4971311", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-vllm-patch-tokenizer-hygon-k100-ai"}
+{"framework": "vllm", "modelAddress": "https://modelscope.cn/models/BAAI/RoboBrain2.5-8B-MT", "modelId": "BAAI/RoboBrain2.5-8B-MT", "submitTime": "2026-09-20T05:16:13.351034+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971313", "taskType": "text-generation", "templateId": "modelhub-live-text-generation-vllm-iluvatar-bi-150"}
diff --git a/manifest.json b/manifest.json
index 3f55c5055c..72fe9cd483 100644
--- a/manifest.json
+++ b/manifest.json
@@ -9,16 +9,16 @@
     ".modelhub_state/queue_cleanup_latest.json": "25b335b72d13b2b3ece533c0d2becb501b6af2b8e22541f81dbe2557eefe8d4f",
     ".modelhub_state/recent_outcomes.jsonl": "a6922b033a6620de4c1eebf27e571bc594980df5d0b7d7d7ae8eb909b24661eb",
     ".modelhub_state/recovery_active_tasks.jsonl": "d8c6b37624bb6cd9e210a8a299e675aefbee3587a13dd0e870c8a1727f42c039",
-    ".modelhub_state/recovery_intents.jsonl": "75f54ac797e5f40872bb5ffc3dfc5faeb31b8650df24f9c9bf57a033ce365e2c",
-    ".modelhub_state/routing_intelligence.json": "97eda7d10b818e9f60e37f84bb11ccba07c9de723b8cd2d067601585d4fe4771",
+    ".modelhub_state/recovery_intents.jsonl": "0435046227a9c9945dbebffdfb0e4f673345448d1bb4413cc293e0b88819fb82",
+    ".modelhub_state/routing_intelligence.json": "5c979addb11f70428e343d11bf3d72db947de2ae79bfc1a17b8ced33f5a118c7",
     ".modelhub_state/submission_exclusions.jsonl": "221be895a524330815b3c5124450b33c94b5f1e68169030c73684bf675d06ec7",
     ".modelhub_state/worker_crashes.jsonl": "a494412048eca6dce83e7284303a6b4b56a445c1274ac201ab8723c739a14f23",
-    "ledger/submissions.jsonl": "924fe2af2db549ffcb0b987735efa7eff311aa92de4d94759fbbc06ef8dce278",
-    "outcomes/submissions.jsonl": "d92833eda8254a2fff0fcdce5d149d5b3da84f8bb3f568379339215d7f3c9b65"
+    "ledger/submissions.jsonl": "07cc2f2d0c1fca21ce2c50706f4c0204038e20dd88ed73c7febba8cd9eb50b8f",
+    "outcomes/submissions.jsonl": "cc820b572b06a5aa16690f20458249a64bf7edada38ad87f0ebdc273b49bbafd"
   },
-  "generation": 10255,
-  "phase": "intent",
+  "generation": 10256,
+  "phase": "result",
   "schemaVersion": 1,
-  "updatedAt": "2026-09-20T05:16:04.379137+00:00",
+  "updatedAt": "2026-09-20T05:16:13.549448+00:00",
   "writerId": "fc715c06e24c4db2ae3e425e435db996"
 }
diff --git a/outcomes/submissions.jsonl b/outcomes/submissions.jsonl
index 84cf8c87fd..09d29ce017 100644
--- a/outcomes/submissions.jsonl
+++ b/outcomes/submissions.jsonl
@@ -1059,3 +1059,21 @@
 {"failReason": null, "framework": "vllm_0_17_0_corex_4_4_0", "lastSyncTime": null, "modelId": "RedHatAI/gemma-4-12B-it-FP8-Dynamic", "modelProfile": {"architectures": ["Gemma4UnifiedForConditionalGeneration"], "configFingerprint": "1cc16a7df03cf7b1485e2fda7e5c05e286a546d04e80bdd595c015004b3852a8", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 15037393456, "estimatedRequiredGiB": 16.842, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "gemma4_unified", "modelscopeFileSize": 15069601989, "modelscopeLicense": "apache-2.0", "modelscopeParams": 12966363184, "modelscopeTags": ["license:apache-2.0", "model_type:gemma4_unified", "library:pytorch", "library:transformer", "library:safetensors", "task:text-generation", "task:image-text-to-text", "custom_tag:fp8", "custom_tag:vllm", "custom_tag:llm-compressor", "custom_tag:compressed-tensors"], "modelscopeTasks": ["text-generation", "image-text-to-text"], "quantizationMethod": "compressed-tensors", "repositoryOnDiskBytes": 15069601989}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T04:46:07.939821+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4970952", "taskType": "text-generation", "verifyResult": null}
 {"failReason": null, "framework": "vllm_0_17_0_corex_4_4_0", "lastSyncTime": null, "modelId": "mlx-community/LFM2.5-1.2B-Thinking-6bit", "modelProfile": {"architectures": ["Lfm2ForCausalLM"], "configFingerprint": "1cc16a7df03cf7b1485e2fda7e5c05e286a546d04e80bdd595c015004b3852a8", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 951093016, "estimatedRequiredGiB": 1.068, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "lfm2", "modelscopeFileSize": 955857175, "modelscopeLicense": "other", "modelscopeParams": 256113408, "modelscopeTags": ["license:other", "model_type:lfm2", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:liquid", "custom_tag:lfm2.5", "custom_tag:edge", "custom_tag:mlx"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 955857175}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:00:40.870810+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971125", "taskType": "text-generation", "verifyResult": null}
 {"failReason": null, "framework": "vllm", "lastSyncTime": null, "modelId": "RedHatAI/gemma-4-12B-it-FP8-Dynamic", "modelProfile": {"architectures": ["Gemma4UnifiedForConditionalGeneration"], "configFingerprint": "d341b8002cf39d3299fd429479e8c8e954583e49d4fec5a640ddc0f8f0cbd054", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 15037393456, "estimatedRequiredGiB": 16.842, "gpuMemoryEvidence": {"memoryGiB": 64.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 64.0, "maximumRepositorySizeGiB": 53.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "gemma4_unified", "modelscopeFileSize": 15069601989, "modelscopeLicense": "apache-2.0", "modelscopeParams": 12966363184, "modelscopeTags": ["license:apache-2.0", "model_type:gemma4_unified", "library:pytorch", "library:transformer", "library:safetensors", "task:text-generation", "task:image-text-to-text", "custom_tag:fp8", "custom_tag:vllm", "custom_tag:llm-compressor", "custom_tag:compressed-tensors"], "modelscopeTasks": ["text-generation", "image-text-to-text"], "quantizationMethod": "compressed-tensors", "repositoryOnDiskBytes": 15069601989}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:01:56.033898+00:00", "targetGpu": "Biren_166m", "taskId": "4971137", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "vllm_fix_tokenizer", "lastSyncTime": null, "modelId": "mlx-community/gemma-4-12B-it-qat-OptiQ-4bit", "modelProfile": {"architectures": ["Gemma4UnifiedForConditionalGeneration"], "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 9003966484, "estimatedRequiredGiB": 10.099, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "gemma4_unified", "modelscopeFileSize": 9036402901, "modelscopeLicense": "apache-2.0", "modelscopeParams": 2463563824, "modelscopeTags": ["license:apache-2.0", "model_type:gemma4_unified", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:mlx", "custom_tag:quantized", "custom_tag:mixed-precision", "custom_tag:4bit", "custom_tag:8bit", "custom_tag:optiq", "custom_tag:qat", "custom_tag:apple-silicon", "custom_tag:text-generation", "custom_tag:image-text-to-text", "custom_tag:gemma-4"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 9036402901}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.156477+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971306", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "vllm_fix_tokenizer", "lastSyncTime": null, "modelId": "mlx-community/KAT-Coder-V2.5-Dev-OptiQ-4bit", "modelProfile": {"architectures": ["Qwen3_5MoeForConditionalGeneration"], "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 21930054875, "estimatedRequiredGiB": 24.532, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5_moe", "modelscopeFileSize": 21950415614, "modelscopeLicense": "apache-2.0", "modelscopeParams": 6024588416, "modelscopeTags": ["license:apache-2.0", "model_type:qwen3_5_moe", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:mlx", "custom_tag:optiq", "custom_tag:code"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 21950415614}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.138764+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971298", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "vllm_fix_tokenizer", "lastSyncTime": null, "modelId": "mlx-community/gemma-4-26B-A4B-it-qat-OptiQ-4bit", "modelProfile": {"architectures": ["Gemma4ForConditionalGeneration"], "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 21855018733, "estimatedRequiredGiB": 24.461, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "gemma4", "modelscopeFileSize": 21887495988, "modelscopeLicense": "apache-2.0", "modelscopeParams": 6144703054, "modelscopeTags": ["license:apache-2.0", "model_type:gemma4", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:mlx", "custom_tag:quantized", "custom_tag:mixed-precision", "custom_tag:4bit", "custom_tag:8bit", "custom_tag:optiq", "custom_tag:qat", "custom_tag:moe", "custom_tag:apple-silicon", "custom_tag:text-generation", "custom_tag:image-text-to-text", "custom_tag:gemma-4"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 21887495988}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.142829+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971299", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "vllm_fix_tokenizer", "lastSyncTime": null, "modelId": "mlx-community/gemma-4-31B-it-qat-OptiQ-4bit", "modelProfile": {"architectures": ["Gemma4ForConditionalGeneration"], "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 23524076260, "estimatedRequiredGiB": 26.327, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "gemma4", "modelscopeFileSize": 23556606333, "modelscopeLicense": "apache-2.0", "modelscopeParams": 6649116524, "modelscopeTags": ["license:apache-2.0", "model_type:gemma4", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:mlx", "custom_tag:quantized", "custom_tag:mixed-precision", "custom_tag:4bit", "custom_tag:8bit", "custom_tag:optiq", "custom_tag:qat", "custom_tag:apple-silicon", "custom_tag:text-generation", "custom_tag:image-text-to-text", "custom_tag:gemma-4"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 23556606333}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.136785+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971300", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "vllm_fix_tokenizer", "lastSyncTime": null, "modelId": "mlx-community/Qwen3.5-122B-A10B-OptiQ-2bit-REAP-63B", "modelProfile": {"architectures": ["Qwen3_5MoeForConditionalGeneration"], "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 25593158704, "estimatedRequiredGiB": 28.626, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5_moe", "modelscopeFileSize": 25613691403, "modelscopeLicense": "apache-2.0", "modelscopeParams": 7626590448, "modelscopeTags": ["license:apache-2.0", "model_type:qwen3_5_moe", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:mlx", "custom_tag:quantized", "custom_tag:expert-pruning", "custom_tag:reap", "custom_tag:moe", "custom_tag:optiq", "custom_tag:apple-silicon", "custom_tag:text-generation"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 25613691403}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.139470+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971297", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "vllm_fix_tokenizer", "lastSyncTime": null, "modelId": "mlx-community/gemma-4-e2b-it-qat-OptiQ-4bit", "modelProfile": {"architectures": ["Gemma4ForConditionalGeneration"], "configFingerprint": "16f74ea0f409861ab0c8169a25d9484cb482e27d6a3ad7c529a70b7cdbfabee3", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 5222073830, "estimatedRequiredGiB": 5.872, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "gemma4", "modelscopeFileSize": 5254590634, "modelscopeLicense": "apache-2.0", "modelscopeParams": 1140034851, "modelscopeTags": ["license:apache-2.0", "model_type:gemma4", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:mlx", "custom_tag:quantized", "custom_tag:mixed-precision", "custom_tag:4bit", "custom_tag:8bit", "custom_tag:optiq", "custom_tag:qat", "custom_tag:apple-silicon", "custom_tag:text-generation", "custom_tag:image-text-to-text", "custom_tag:gemma-4"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 5254590634}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.160907+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971308", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "transformers", "lastSyncTime": null, "modelId": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 20683241105, "estimatedRequiredGiB": 23.138, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": 20703487237, "modelscopeLicense": "apache-2.0", "modelscopeParams": 6086364400, "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: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", "custom_tag:qwen3-8", "custom_tag:qwen-3.8", "custom_tag:local-llm", "custom_tag:llm", "custom_tag:m5", "custom_tag:m5-max", "custom_tag:m4", "custom_tag:m3", "custom_tag:macbook-pro", "custom_tag:mac-studio", "custom_tag:opencode", "custom_tag:claude-code", "custom_tag:27b", "custom_tag:qwen3.8-27b", "custom_tag:qwen3-8-27b"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 20703487237}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.144215+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971303", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "transformers", "lastSyncTime": null, "modelId": "mlx-community/gemma-4-e4b-it-OptiQ-4bit", "modelProfile": {"architectures": ["Gemma4ForConditionalGeneration"], "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 7486327132, "estimatedRequiredGiB": 8.403, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "gemma4", "modelscopeFileSize": 7518908360, "modelscopeLicense": "gemma", "modelscopeParams": 2227418442, "modelscopeTags": ["license:gemma", "model_type:gemma4", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:mlx", "custom_tag:quantized", "custom_tag:mixed-precision", "custom_tag:4bit", "custom_tag:8bit", "custom_tag:optiq", "custom_tag:apple-silicon", "custom_tag:text-generation", "custom_tag:gemma-4"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 7518908360}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.158936+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971302", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "transformers", "lastSyncTime": null, "modelId": "Youssofal/Qwen3.8-27B-MTPLX-Optimized-Speed-FP16", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 20683239637, "estimatedRequiredGiB": 23.138, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": 20703971805, "modelscopeLicense": "apache-2.0", "modelscopeParams": 6086364400, "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", "custom_tag:qwen3-8", "custom_tag:qwen-3.8", "custom_tag:local-llm", "custom_tag:llm", "custom_tag:m5", "custom_tag:m5-max", "custom_tag:m4", "custom_tag:m3", "custom_tag:macbook-pro", "custom_tag:mac-studio", "custom_tag:opencode", "custom_tag:claude-code", "custom_tag:27b", "custom_tag:qwen3.8-27b", "custom_tag:qwen3-8-27b"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 20703971805}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.154796+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971307", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "transformers", "lastSyncTime": null, "modelId": "Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 16293474987, "estimatedRequiredGiB": 18.232, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": 16313701503, "modelscopeLicense": "apache-2.0", "modelscopeParams": 4665462000, "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: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:chat", "custom_tag:qwen3-8", "custom_tag:qwen-3.8", "custom_tag:local-llm", "custom_tag:llm", "custom_tag:m5", "custom_tag:m5-max", "custom_tag:m4", "custom_tag:m3", "custom_tag:macbook-pro", "custom_tag:mac-studio", "custom_tag:opencode", "custom_tag:claude-code", "custom_tag:27b", "custom_tag:qwen3.8-27b", "custom_tag:qwen3-8-27b", "custom_tag:coding"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 16313701503}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.140867+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971304", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "transformers", "lastSyncTime": null, "modelId": "mlx-community/gemma-4-e2b-it-OptiQ-4bit", "modelProfile": {"architectures": ["Gemma4ForConditionalGeneration"], "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 5226595850, "estimatedRequiredGiB": 5.878, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "gemma4", "modelscopeFileSize": 5259113427, "modelscopeLicense": "gemma", "modelscopeParams": 1141165347, "modelscopeTags": ["license:gemma", "model_type:gemma4", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:mlx", "custom_tag:quantized", "custom_tag:mixed-precision", "custom_tag:4bit", "custom_tag:8bit", "custom_tag:optiq", "custom_tag:apple-silicon", "custom_tag:text-generation", "custom_tag:gemma-4"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 5259113427}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.149648+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971305", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "transformers", "lastSyncTime": null, "modelId": "Arain119/Sophia", "modelProfile": {"architectures": ["SophiaForCausalLM"], "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 2226652224, "estimatedRequiredGiB": 7.28, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "sophia_hybrid", "modelscopeFileSize": 6513869659, "modelscopeLicense": "Apache License 2.0", "modelscopeParams": 1113293776, "modelscopeTags": ["license:Apache License 2.0", "model_type:sophia_hybrid", "library:transformer", "library:safetensors", "library:pytorch", "task:text-generation"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 6513869659}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.151906+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971301", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "transformers", "lastSyncTime": null, "modelId": "Youssofal/Qwen3.8-27B-MTPLX-Bare-Speed-FP16", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 16293473597, "estimatedRequiredGiB": 18.233, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": 16314185171, "modelscopeLicense": "apache-2.0", "modelscopeParams": 4665462000, "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:chat", "custom_tag:qwen3-8", "custom_tag:qwen-3.8", "custom_tag:local-llm", "custom_tag:llm", "custom_tag:m5", "custom_tag:m5-max", "custom_tag:m4", "custom_tag:m3", "custom_tag:macbook-pro", "custom_tag:mac-studio", "custom_tag:opencode", "custom_tag:claude-code", "custom_tag:27b", "custom_tag:qwen3.8-27b", "custom_tag:qwen3-8-27b", "custom_tag:coding"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 16314185171}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.300793+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971309", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "transformers", "lastSyncTime": null, "modelId": "mlx-community/Devstral-Small-2-24B-Instruct-2512-OptiQ-4bit", "modelProfile": {"architectures": ["Mistral3ForConditionalGeneration"], "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 17367755740, "estimatedRequiredGiB": 19.429, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "mistral3", "modelscopeFileSize": 17385072379, "modelscopeLicense": "apache-2.0", "modelscopeParams": 4929899520, "modelscopeTags": ["license:apache-2.0", "model_type:mistral3", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:mlx", "custom_tag:quantized", "custom_tag:mixed-precision", "custom_tag:4bit", "custom_tag:8bit", "custom_tag:optiq", "custom_tag:apple-silicon", "custom_tag:text-generation", "custom_tag:mistral", "custom_tag:mistral3", "custom_tag:devstral"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 17385072379}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.323221+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971310", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "transformers", "lastSyncTime": null, "modelId": "mlx-community/LFM2.5-1.2B-Instruct-4bit", "modelProfile": {"architectures": ["Lfm2ForCausalLM"], "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 658540250, "estimatedRequiredGiB": 0.741, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "lfm2", "modelscopeFileSize": 663396475, "modelscopeLicense": "other", "modelscopeParams": 182975232, "modelscopeTags": ["license:other", "model_type:lfm2", "library:mlx", "library:safetensors", "library:pytorch", "task:text-generation", "custom_tag:liquid", "custom_tag:lfm2.5", "custom_tag:edge", "custom_tag:mlx"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 663396475}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.330122+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971312", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "transformers", "lastSyncTime": null, "modelId": "Youssofal/Qwen3.5-9B-MTPLX-Optimized-Speed", "modelProfile": {"architectures": ["Qwen3_5ForConditionalGeneration"], "configFingerprint": "3086a71adcdca90005b4967eb30749c44cd82389aff8d599758e4b91675d40dc", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 8674988799, "estimatedRequiredGiB": 9.718, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_5", "modelscopeFileSize": 8695119291, "modelscopeLicense": "apache-2.0", "modelscopeParams": 2415484144, "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:speculative-decoding", "custom_tag:qwen", "custom_tag:qwen3", "custom_tag:mtp", "custom_tag:mtplx", "custom_tag:local-ai", "custom_tag:qwen3-5", "custom_tag:9b", "custom_tag:6-bit"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 8695119291}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.358703+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971314", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "vllm-patch-tokenizer", "lastSyncTime": null, "modelId": "RedHatAI/gemma-4-12B-it-FP8-Dynamic", "modelProfile": {"architectures": ["Gemma4UnifiedForConditionalGeneration"], "configFingerprint": "0d5bc82787a89d4313672808d1a8629a1fc675c91aa6a08865f770aa24447fbc", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 15037393456, "estimatedRequiredGiB": 16.842, "gpuMemoryEvidence": {"memoryGiB": 64.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 64.0, "maximumRepositorySizeGiB": 53.333, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "gemma4_unified", "modelscopeFileSize": 15069601989, "modelscopeLicense": "apache-2.0", "modelscopeParams": 12966363184, "modelscopeTags": ["license:apache-2.0", "model_type:gemma4_unified", "library:pytorch", "library:transformer", "library:safetensors", "task:text-generation", "task:image-text-to-text", "custom_tag:fp8", "custom_tag:vllm", "custom_tag:llm-compressor", "custom_tag:compressed-tensors"], "modelscopeTasks": ["text-generation", "image-text-to-text"], "quantizationMethod": "compressed-tensors", "repositoryOnDiskBytes": 15069601989}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.324611+00:00", "targetGpu": "hygon_k100-ai", "taskId": "4971311", "taskType": "text-generation", "verifyResult": null}
+{"failReason": null, "framework": "vllm", "lastSyncTime": null, "modelId": "BAAI/RoboBrain2.5-8B-MT", "modelProfile": {"architectures": ["Qwen3VLForConditionalGeneration"], "configFingerprint": "ac4b2845ddd4bdce478dd2b65134ee3ab3d259a119afe6841bf068b135fd4892", "configOptimization": {"applied": false, "source": "official"}, "configSource": "modelhub_live", "estimatedLoadBytes": 17534339552, "estimatedRequiredGiB": 19.609, "gpuMemoryEvidence": {"memoryGiB": 32.0, "source": "local_modelhub_preflight_oom"}, "gpuMemoryGiB": 32.0, "maximumRepositorySizeGiB": 26.667, "memorySizingBasis": "recursive_repository_on_disk", "modelCard": {}, "modelType": "qwen3_vl", "modelscopeFileSize": 17545918174, "modelscopeLicense": "apache-2.0", "modelscopeParams": 8767123696, "modelscopeTags": ["license:apache-2.0", "model_type:qwen3_vl", "library:safetensors", "library:pytorch", "task:text-generation"], "modelscopeTasks": ["text-generation"], "quantizationMethod": null, "repositoryOnDiskBytes": 17545918174}, "outcome": "pending", "status": "pending", "submitTime": "2026-09-20T05:16:13.351034+00:00", "targetGpu": "Iluvatar_bi-150", "taskId": "4971313", "taskType": "text-generation", "verifyResult": null}