lvkaokao commited on
Commit
dfb4059
·
verified ·
1 Parent(s): b08dbed

Add google/gemma-4-31B-it to quant queue

Browse files
google/gemma-4-31B-it_quant_request_False_W4A16_4bit_int4_RTN_20260622T033748Z.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"job_type": "quantization & evaluation", "model": "google/gemma-4-31B-it", "revision": "main", "private": false, "params": 31.27, "architectures": "Gemma4ForConditionalGeneration", "num_layers": 60, "input_dtype": "bfloat16", "input_bits": 16, "model_params": 31.27, "model_weight_gb": 62.55, "quant_scheme": "INT4 (W4A16)", "quant_precision": "4bit", "quant_weight_dtype": "int4", "quant_bits": 4, "quant_memory_gb": 3.13, "quant_gpu_type": "NVIDIA GeForce RTX 4090", "quant_gpu_nums": 1, "eval_memory_gb": 34.4, "eval_gpu_type": "NVIDIA RTX PRO 6000 Blackwell Server Edition", "eval_gpu_nums": 1, "quant_model_size_gb": 15.63, "hardware": "NVIDIA RTX PRO 6000 Blackwell Server Edition", "gpu_nums": 1, "method": "RTN", "script": "auto_quant", "status": "Pending", "submitted_time": "2026-06-22T03:37:48Z", "submitted_by": "wenjiao", "submitted_orgs": ["Intel"], "model_type": "quantization", "job_id": -1, "job_start_time": null, "likes": 3043, "license": "apache-2.0", "still_on_hub": true, "tags": ["moe"], "downloads": 0, "created_at": ""}