add AIBOM

microsoft_MAI-DS-R1.json

@@ -0,0 +1,66 @@
+{
+    "bomFormat": "CycloneDX",
+    "specVersion": "1.6",
+    "serialNumber": "urn:uuid:0c1f0396-4043-4195-820e-4870a03c00ac",
+    "version": 1,
+    "metadata": {
+        "timestamp": "2025-06-05T09:40:31.549137+00:00",
+        "component": {
+            "type": "machine-learning-model",
+            "bom-ref": "microsoft/MAI-DS-R1-8d11dfdf-fb8a-57f6-bde0-49aa8aab572c",
+            "name": "microsoft/MAI-DS-R1",
+            "externalReferences": [
+                {
+                    "url": "https://huggingface.co/microsoft/MAI-DS-R1",
+                    "type": "documentation"
+                }
+            ],
+            "modelCard": {
+                "modelParameters": {
+                    "task": "text-generation",
+                    "architectureFamily": "deepseek_v3",
+                    "modelArchitecture": "DeepseekV3ForCausalLM"
+                },
+                "properties": [
+                    {
+                        "name": "library_name",
+                        "value": "transformers"
+                    },
+                    {
+                        "name": "base_model",
+                        "value": "deepseek-ai/DeepSeek-R1"
+                    }
+                ]
+            },
+            "authors": [
+                {
+                    "name": "microsoft"
+                }
+            ],
+            "licenses": [
+                {
+                    "license": {
+                        "id": "MIT",
+                        "url": "https://spdx.org/licenses/MIT.html"
+                    }
+                }
+            ],
+            "description": "MAI-DS-R1 is a DeepSeek-R1 reasoning model that has been post-trained by Microsoft AI team to fill in information gaps in the previous version of the model and to improve its risk profile, while maintaining R1 reasoning capabilities. The model was trained using 110k Safety and Non-Compliance examples from [Tulu](https://huggingface.co/datasets/allenai/tulu-3-sft-mixture) 3 SFT dataset, in addition to a dataset of ~350k multilingual examples internally developed capturing various topics with reported biases.MAI-DS-R1 has successfully unblocked the majority of previously blocked queries from the original R1 model while outperforming the recently published R1-1776 model (post-trained by Perplexity) in relevant safety benchmarks.  These results were achieved while preserving the general reasoning capabilities of the original DeepSeek-R1.*Please note: Microsoft has post-trained this model to address certain limitations relevant to its outputs, but previous limitations and considerations for the model remain, including security considerations.*",
+            "tags": [
+                "transformers",
+                "safetensors",
+                "deepseek_v3",
+                "text-generation",
+                "conversational",
+                "custom_code",
+                "base_model:deepseek-ai/DeepSeek-R1",
+                "base_model:finetune:deepseek-ai/DeepSeek-R1",
+                "license:mit",
+                "autotrain_compatible",
+                "text-generation-inference",
+                "endpoints_compatible",
+                "region:us"
+            ]
+        }
+    }
+}