Add llmfan46/Qwen3.6-35B-A3B-uncensored-heretic to quant queue
Browse files
llmfan46/Qwen3.6-35B-A3B-uncensored-heretic_quant_request_False_MXFP8_8bit_mxfp8_TUNING_20260710T083922Z.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"job_type": "quantization & evaluation", "model": "llmfan46/Qwen3.6-35B-A3B-uncensored-heretic", "revision": "main", "private": false, "params": 35.11, "architectures": "Qwen3_5MoeForConditionalGeneration", "num_layers": 40, "input_dtype": "bfloat16", "input_bits": 16, "model_params": 35.11, "model_weight_gb": 70.21, "quant_scheme": "MXFP8", "quant_precision": "8bit", "quant_weight_dtype": "mxfp8", "quant_bits": 8, "quant_memory_gb": 5.27, "quant_gpu_type": "AWS B200", "quant_gpu_nums": 1, "eval_memory_gb": 77.24, "eval_gpu_type": "AWS B200", "eval_gpu_nums": 1, "quant_model_size_gb": 35.11, "hardware": "AWS B200", "gpu_nums": 1, "method": "TUNING", "script": "auto_quant", "status": "Pending", "submitted_time": "2026-07-10T08:39:22Z", "submitted_by": "lvkaokao", "submitted_orgs": ["Intel", "INCModel", "INCModel2", "INCModelSharing"], "model_type": "quantization", "job_id": -1, "job_start_time": null, "export_format": "llm_compressor", "ignore_layers": "lm_head,mlp.gate,language_model.visual.", "layer_config": "{mlp.experts:{bits:4,data_type:mx_fp}}", "agent_type": "local", "cuda_visible_devices": "4", "likes": 86, "license": "apache-2.0", "still_on_hub": true, "tags": ["moe"], "downloads": 0, "created_at": ""}
|