Upload deploy.json with huggingface_hub
Browse files- deploy.json +13 -4
deploy.json
CHANGED
|
@@ -1,14 +1,23 @@
|
|
| 1 |
{
|
| 2 |
"parameters": {
|
| 3 |
-
"max_input_length": 1024,
|
| 4 |
"max_total_tokens": 2048,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
"max_batch_size": 1,
|
| 6 |
-
"
|
| 7 |
},
|
| 8 |
"hardware": {
|
|
|
|
| 9 |
"accelerator": "gpu",
|
| 10 |
-
"
|
|
|
|
|
|
|
| 11 |
},
|
| 12 |
"framework_type": "pytorch",
|
| 13 |
-
"
|
|
|
|
|
|
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"parameters": {
|
|
|
|
| 3 |
"max_total_tokens": 2048,
|
| 4 |
+
"max_input_length": 1024,
|
| 5 |
+
"max_batch_total_tokens": 8192,
|
| 6 |
+
"max_concurrent_requests": 1,
|
| 7 |
+
"max_best_of": 1,
|
| 8 |
+
"max_stop_sequences": 4,
|
| 9 |
"max_batch_size": 1,
|
| 10 |
+
"waiting_served_ratio": 1.2
|
| 11 |
},
|
| 12 |
"hardware": {
|
| 13 |
+
"task_type": "text-generation",
|
| 14 |
"accelerator": "gpu",
|
| 15 |
+
"num_gpus": 1,
|
| 16 |
+
"gpu_memory_gb": 24,
|
| 17 |
+
"distributed_setup": false
|
| 18 |
},
|
| 19 |
"framework_type": "pytorch",
|
| 20 |
+
"torch_compile": false,
|
| 21 |
+
"trust_remote_code": true,
|
| 22 |
+
"disable_custom_kernels": true
|
| 23 |
}
|