Add stepfun-ai/Step-3.7-Flash to quant queue
Browse files
stepfun-ai/Step-3.7-Flash_quant_request_False_W4A16_4bit_int4_RTN.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"job_type": "quantization & evaluation", "model": "stepfun-ai/Step-3.7-Flash", "revision": "main", "private": false, "params": 201.37, "architectures": "Step3p7ForConditionalGeneration", "num_layers": 45, "input_dtype": "bfloat16", "input_bits": 16, "model_params": 201.37, "model_weight_gb": 402.73, "quant_scheme": "INT4 (W4A16)", "quant_precision": "4bit", "quant_weight_dtype": "int4", "quant_bits": 4, "quant_memory_gb": 26.85, "quant_gpu_type": "NVIDIA GeForce RTX 5090", "quant_gpu_nums": 1, "eval_memory_gb": 221.51, "eval_gpu_type": "NVIDIA H200", "eval_gpu_nums": 2, "quant_model_size_gb": 100.69, "hardware": "NVIDIA H200", "gpu_nums": 2, "method": "RTN", "script": "auto_quant", "status": "Pending", "submitted_time": "2026-08-03T09:29:48Z", "submitted_by": "wenjiao", "submitted_orgs": ["Intel"], "model_type": "quantization", "job_id": -1, "job_start_time": null, "likes": 433, "license": "apache-2.0", "still_on_hub": true, "tags": ["moe"], "downloads": 0, "created_at": ""}
|