docs: update the naming convention of customizer configs

NeMo Customizer now uses configs that reference GPU memomry instead of SKU.

Author: gabwow

nemo/data-flywheel/embedding-finetuning/2_finetuning_and_inference.ipynb

@@ -526,7 +526,7 @@
     "A customization configuration defines the model, hardware, and training settings for fine-tuning jobs.\n",
     "\n",
     "**Off-the-Shelf vs Custom Configurations:**\n",
-    "- **Off-the-shelf configs** (e.g., `llama-3.2-1b-embed@v1.0.0+A100`) are pre-built and ready to use. To use one, you would reference it by name instead of creating a new config.\n",
+    "- **Off-the-shelf configs** (e.g., `llama-3.2-nv-embedqa-1b@v2+80GB`) are pre-built and ready to use. To use one, you would reference it by name instead of creating a new config.\n",
     "- **Custom configs** let you specify your own training parameters, hardware requirements, and model settings.\n",
     "\n",
     "**The `target` Parameter:** Specifies the base model checkpoint to fine-tune. We're using [llama-3_2-nv-embedqa-1b-v2](https://build.nvidia.com/nvidia/llama-3_2-nv-embedqa-1b-v2), a multilingual embedding model trained for text question-answering retrieval tasks.\n",
@@ -1612,4 +1612,4 @@
  },
  "nbformat": 4,
  "nbformat_minor": 5
-}
+}

nemo/data-flywheel/tool-calling/2_finetuning_and_inference.ipynb

@@ -93,7 +93,7 @@
       "Entity Store, Customizer, Evaluator endpoint: http://nemo.test\n",
       "NIM endpoint: http://nim.test\n",
       "Namespace: xlam-tutorial-ns\n",
-      "Base Model for Customization: meta/llama-3.2-1b-instruct@v1.0.0+A100\n"
+      "Base Model for Customization: meta/llama-3.2-1b-instruct@v1.0.0+80GB\n"
      ]
     }
    ],
@@ -576,7 +576,7 @@
     {
      "data": {
       "text/plain": [
-       "CustomizationJobOutput(config='meta/llama-3.2-1b-instruct@v1.0.0+A100', config_snapshot=ConfigSnapshot(base_model='meta/llama-3.2-1b-instruct', max_seq_length=4096, precision='bf16-mixed', training_option=CustomizationTrainingOption(finetuning_type='lora', micro_batch_size=1, num_gpus=1, training_type='sft', data_parallel_size=1, num_nodes=1, pipeline_parallel_size=1, tensor_parallel_size=1, use_sequence_parallel=False), chat_prompt_template=None, name=None, namespace=None, prompt_template='{prompt} {completion}', tokenizer=None), dataset='xlam-tutorial-ns/xlam-ft-dataset', hyperparameters=Hyperparameters(finetuning_type='lora', batch_size=16, distillation=None, epochs=2, learning_rate=0.0001, log_every_n_steps=None, lora=Lora(adapter_dim=32, adapter_dropout=0.1, alpha=16, target_modules=None), p_tuning=None, sequence_packing_enabled=False, sft=None, training_type='sft', val_check_interval=None, weight_decay=None), id='cust-FarcM8gwhL1XFDXQ57qGLL', created_at=datetime.datetime(2025, 6, 20, 4, 20, 21, 387984), dataset_parameters=None, description=None, integrations=None, namespace='default', output_model='xlam-tutorial-ns/llama-3.2-1b-xlam-run1@cust-FarcM8gwhL1XFDXQ57qGLL', ownership=None, project=None, status='created', status_details={'created_at': '2025-06-20T04:20:22.061480', 'updated_at': '2025-06-20T04:20:22.061480', 'elapsed_time': 0.0, 'steps_completed': 0, 'epochs_completed': 0, 'percentage_done': 0.0, 'status_logs': [{'updated_at': '2025-06-20T04:20:22.061480', 'message': 'created'}]}, updated_at=datetime.datetime(2025, 6, 20, 4, 20, 21, 387989), warnings=None)"
+       "CustomizationJobOutput(config='meta/llama-3.2-1b-instruct@v1.0.0+80GB', config_snapshot=ConfigSnapshot(base_model='meta/llama-3.2-1b-instruct', max_seq_length=4096, precision='bf16-mixed', training_option=CustomizationTrainingOption(finetuning_type='lora', micro_batch_size=1, num_gpus=1, training_type='sft', data_parallel_size=1, num_nodes=1, pipeline_parallel_size=1, tensor_parallel_size=1, use_sequence_parallel=False), chat_prompt_template=None, name=None, namespace=None, prompt_template='{prompt} {completion}', tokenizer=None), dataset='xlam-tutorial-ns/xlam-ft-dataset', hyperparameters=Hyperparameters(finetuning_type='lora', batch_size=16, distillation=None, epochs=2, learning_rate=0.0001, log_every_n_steps=None, lora=Lora(adapter_dim=32, adapter_dropout=0.1, alpha=16, target_modules=None), p_tuning=None, sequence_packing_enabled=False, sft=None, training_type='sft', val_check_interval=None, weight_decay=None), id='cust-FarcM8gwhL1XFDXQ57qGLL', created_at=datetime.datetime(2025, 6, 20, 4, 20, 21, 387984), dataset_parameters=None, description=None, integrations=None, namespace='default', output_model='xlam-tutorial-ns/llama-3.2-1b-xlam-run1@cust-FarcM8gwhL1XFDXQ57qGLL', ownership=None, project=None, status='created', status_details={'created_at': '2025-06-20T04:20:22.061480', 'updated_at': '2025-06-20T04:20:22.061480', 'elapsed_time': 0.0, 'steps_completed': 0, 'epochs_completed': 0, 'percentage_done': 0.0, 'status_logs': [{'updated_at': '2025-06-20T04:20:22.061480', 'message': 'created'}]}, updated_at=datetime.datetime(2025, 6, 20, 4, 20, 21, 387989), warnings=None)"
       ]
      },
      "execution_count": 56,

nemo/data-flywheel/tool-calling/config.py

@@ -18,7 +18,7 @@
 
 # (Optional) Configure the base model. Must be one supported by the NeMo Customizer deployment!
 BASE_MODEL = "meta/llama-3.2-1b-instruct"
-BASE_MODEL_VERSION = "v1.0.0+A100"
+BASE_MODEL_VERSION = "v1.0.0+80GB"
 
 # (Optional) Configure the custom model. Ensure the custom model name can be pass to the other notebooks
 CUSTOM_MODEL = f"{NMS_NAMESPACE}/llama-3.2-1b-xlam-run1@v1"