add AIBOM

Dear model owner(s),
We are a group of researchers investigating the usefulness of sharing AIBOMs (Artificial Intelligence Bill of Materials) to document AI models – AIBOMs are machine-readable structured lists of components (e.g., datasets and models) used to enhance transparency in AI-model supply chains.

To pursue the above-mentioned objective, we identified popular models on HuggingFace and, based on your model card (and some configuration information available in HuggingFace), we generated your AIBOM according to the CyclonDX (v1.6) standard (see https://cyclonedx.org/docs/1.6/json/). AIBOMs are generated as JSON files by using the following open-source supporting tool: https://github.com/MSR4SBOM/ALOHA (technical details are available in the research paper: https://github.com/MSR4SBOM/ALOHA/blob/main/ALOHA.pdf).

The JSON file in this pull request is your AIBOM (see https://github.com/MSR4SBOM/ALOHA/blob/main/documentation.json for details on its structure).

Clearly, the submitted AIBOM matches the current model information, yet it can be easily regenerated when the model evolves, using the aforementioned AIBOM generator tool.

We open this pull request containing an AIBOM of your AI model, and hope it will be considered. We would also like to hear your opinion on the usefulness (or not) of AIBOM by answering a 3-minute anonymous survey: https://forms.gle/WGffSQD5dLoWttEe7.

Thanks in advance, and regards,
Riccardo D’Avino, Fatima Ahmed, Sabato Nocera, Simone Romano, Giuseppe Scanniello (University of Salerno, Italy),
Massimiliano Di Penta (University of Sannio, Italy),
The MSR4SBOM team

Files changed (1) hide show

openchat_openchat_3.5.json +526 -0

openchat_openchat_3.5.json ADDED Viewed

	@@ -0,0 +1,526 @@

+{
+    "bomFormat": "CycloneDX",
+    "specVersion": "1.6",
+    "serialNumber": "urn:uuid:47a63261-3969-4c57-9e56-fb97f0d285a2",
+    "version": 1,
+    "metadata": {
+        "timestamp": "2025-06-05T09:41:28.741689+00:00",
+        "component": {
+            "type": "machine-learning-model",
+            "bom-ref": "openchat/openchat_3.5-7973a308-bdbf-59a9-9f73-be9d9f2b78c3",
+            "name": "openchat/openchat_3.5",
+            "externalReferences": [
+                {
+                    "url": "https://huggingface.co/openchat/openchat_3.5",
+                    "type": "documentation"
+                }
+            ],
+            "modelCard": {
+                "modelParameters": {
+                    "task": "text-generation",
+                    "architectureFamily": "mistral",
+                    "modelArchitecture": "MistralForCausalLM",
+                    "datasets": [
+                        {
+                            "ref": "openchat/openchat_sharegpt4_dataset-79a9a029-636a-5348-a67b-03818aca7f78"
+                        },
+                        {
+                            "ref": "imone/OpenOrca_FLAN-56466c9b-33bb-5f23-ab7c-388f28f07644"
+                        },
+                        {
+                            "ref": "LDJnr/LessWrong-Amplify-Instruct-fd396583-6aa0-5026-9748-4409738f0812"
+                        },
+                        {
+                            "ref": "LDJnr/Pure-Dove-e5138f01-f77d-5510-9285-df11d660b143"
+                        },
+                        {
+                            "ref": "LDJnr/Verified-Camel-6c954918-6150-534c-952b-ea2cff17c0b2"
+                        },
+                        {
+                            "ref": "tiedong/goat-38642401-79d7-541d-a92f-d3d7acdb8db8"
+                        },
+                        {
+                            "ref": "glaiveai/glaive-code-assistant-4cf9eb44-c4b0-5c31-b48c-ce95c2926a6d"
+                        },
+                        {
+                            "ref": "meta-math/MetaMathQA-c6cf810a-8b06-5552-a876-53681c5fe9a1"
+                        },
+                        {
+                            "ref": "OpenAssistant/oasst_top1_2023-08-25-10cff0b1-2289-5b35-8ea8-59253cea318a"
+                        },
+                        {
+                            "ref": "TIGER-Lab/MathInstruct-9d9c997d-f6c1-5029-96fd-6003c4f0ec06"
+                        }
+                    ]
+                },
+                "properties": [
+                    {
+                        "name": "library_name",
+                        "value": "transformers"
+                    }
+                ]
+            },
+            "authors": [
+                {
+                    "name": "openchat"
+                }
+            ],
+            "licenses": [
+                {
+                    "license": {
+                        "id": "Apache-2.0",
+                        "url": "https://spdx.org/licenses/Apache-2.0.html"
+                    }
+                }
+            ],
+            "tags": [
+                "transformers",
+                "pytorch",
+                "mistral",
+                "text-generation",
+                "openchat",
+                "C-RLFT",
+                "conversational",
+                "dataset:openchat/openchat_sharegpt4_dataset",
+                "dataset:imone/OpenOrca_FLAN",
+                "dataset:LDJnr/LessWrong-Amplify-Instruct",
+                "dataset:LDJnr/Pure-Dove",
+                "dataset:LDJnr/Verified-Camel",
+                "dataset:tiedong/goat",
+                "dataset:glaiveai/glaive-code-assistant",
+                "dataset:meta-math/MetaMathQA",
+                "dataset:OpenAssistant/oasst_top1_2023-08-25",
+                "dataset:TIGER-Lab/MathInstruct",
+                "arxiv:2309.11235",
+                "arxiv:2303.08774",
+                "license:apache-2.0",
+                "autotrain_compatible",
+                "text-generation-inference",
+                "endpoints_compatible",
+                "region:us"
+            ]
+        }
+    },
+    "components": [
+        {
+            "type": "data",
+            "bom-ref": "openchat/openchat_sharegpt4_dataset-79a9a029-636a-5348-a67b-03818aca7f78",
+            "name": "openchat/openchat_sharegpt4_dataset",
+            "data": [
+                {
+                    "type": "dataset",
+                    "bom-ref": "openchat/openchat_sharegpt4_dataset-79a9a029-636a-5348-a67b-03818aca7f78",
+                    "name": "openchat/openchat_sharegpt4_dataset",
+                    "contents": {
+                        "url": "https://huggingface.co/datasets/openchat/openchat_sharegpt4_dataset",
+                        "properties": [
+                            {
+                                "name": "task_categories",
+                                "value": "conversational, text-generation"
+                            },
+                            {
+                                "name": "language",
+                                "value": "en"
+                            },
+                            {
+                                "name": "size_categories",
+                                "value": "1K<n<10K"
+                            },
+                            {
+                                "name": "pretty_name",
+                                "value": "OpenChat"
+                            }
+                        ]
+                    },
+                    "governance": {
+                        "owners": [
+                            {
+                                "organization": {
+                                    "name": "openchat",
+                                    "url": "https://huggingface.co/openchat"
+                                }
+                            }
+                        ]
+                    },
+                    "description": "This repository contains cleaned and filtered ShareGPT GPT-4 data used to train OpenChat. Details can be found in the OpenChat repository.\n"
+                }
+            ]
+        },
+        {
+            "type": "data",
+            "bom-ref": "imone/OpenOrca_FLAN-56466c9b-33bb-5f23-ab7c-388f28f07644",
+            "name": "imone/OpenOrca_FLAN",
+            "data": [
+                {
+                    "type": "dataset",
+                    "bom-ref": "imone/OpenOrca_FLAN-56466c9b-33bb-5f23-ab7c-388f28f07644",
+                    "name": "imone/OpenOrca_FLAN",
+                    "contents": {
+                        "url": "https://huggingface.co/datasets/imone/OpenOrca_FLAN",
+                        "properties": [
+                            {
+                                "name": "license",
+                                "value": "mit"
+                            }
+                        ]
+                    },
+                    "governance": {
+                        "owners": [
+                            {
+                                "organization": {
+                                    "name": "imone",
+                                    "url": "https://huggingface.co/imone"
+                                }
+                            }
+                        ]
+                    },
+                    "description": "This is the OpenOrca GPT4 subset with the original FLAN answers. Each even row (indexed starting from 0) contains the OpenOrca GPT4 answer, while each odd row contains the corresponding FLAN answer.\n"
+                }
+            ]
+        },
+        {
+            "type": "data",
+            "bom-ref": "LDJnr/LessWrong-Amplify-Instruct-fd396583-6aa0-5026-9748-4409738f0812",
+            "name": "LDJnr/LessWrong-Amplify-Instruct",
+            "data": [
+                {
+                    "type": "dataset",
+                    "bom-ref": "LDJnr/LessWrong-Amplify-Instruct-fd396583-6aa0-5026-9748-4409738f0812",
+                    "name": "LDJnr/LessWrong-Amplify-Instruct",
+                    "contents": {
+                        "url": "https://huggingface.co/datasets/LDJnr/LessWrong-Amplify-Instruct",
+                        "properties": [
+                            {
+                                "name": "task_categories",
+                                "value": "conversational, question-answering, text-generation"
+                            },
+                            {
+                                "name": "language",
+                                "value": "en"
+                            },
+                            {
+                                "name": "size_categories",
+                                "value": "n<1K"
+                            },
+                            {
+                                "name": "pretty_name",
+                                "value": "LessWrong-Amplify-Instruct"
+                            },
+                            {
+                                "name": "license",
+                                "value": "apache-2.0"
+                            }
+                        ]
+                    },
+                    "governance": {
+                        "owners": [
+                            {
+                                "organization": {
+                                    "name": "LDJnr",
+                                    "url": "https://huggingface.co/LDJnr"
+                                }
+                            }
+                        ]
+                    },
+                    "description": "\n\t\n\t\t\n\t\tThis is the Official LessWrong-Amplify-Instruct dataset. Over 500 multi-turn examples, and many more coming soon!\n\t\n\n\nThis leverages Amplify-Instruct method to extend thousands of scraped Less-Wrong posts into advanced in-depth multi-turn conversations.\n\nComprised of over 500 highly filtered multi-turn synthetic conversations.\n\nAverage context length per conversation is over 2,000 tokens. (will measure this more accurately soon)\n\nSynthetically created using a newly developed pipeline\u2026 See the full description on the dataset page: https://huggingface.co/datasets/LDJnr/LessWrong-Amplify-Instruct."
+                }
+            ]
+        },
+        {
+            "type": "data",
+            "bom-ref": "LDJnr/Pure-Dove-e5138f01-f77d-5510-9285-df11d660b143",
+            "name": "LDJnr/Pure-Dove",
+            "data": [
+                {
+                    "type": "dataset",
+                    "bom-ref": "LDJnr/Pure-Dove-e5138f01-f77d-5510-9285-df11d660b143",
+                    "name": "LDJnr/Pure-Dove",
+                    "contents": {
+                        "url": "https://huggingface.co/datasets/LDJnr/Pure-Dove",
+                        "properties": [
+                            {
+                                "name": "task_categories",
+                                "value": "conversational, question-answering, text-generation"
+                            },
+                            {
+                                "name": "language",
+                                "value": "en"
+                            },
+                            {
+                                "name": "size_categories",
+                                "value": "1K<n<10K"
+                            },
+                            {
+                                "name": "pretty_name",
+                                "value": "Pure-Dove"
+                            },
+                            {
+                                "name": "license",
+                                "value": "apache-2.0"
+                            }
+                        ]
+                    },
+                    "governance": {
+                        "owners": [
+                            {
+                                "organization": {
+                                    "name": "LDJnr",
+                                    "url": "https://huggingface.co/LDJnr"
+                                }
+                            }
+                        ]
+                    },
+                    "description": "\n\t\n\t\t\n\t\tThis is the Official Pure-Dove dataset. Over 3K multi-turn examples, and many more coming soon!\n\t\n\nThis dataset aims to be the largest highest quality cluster of real human back and forth conversations with GPT-4.\nSteps have even been done to ensure that only the best GPT-4 conversations in comparisons are kept, there are many instances where two GPT-4 responses are rated as equal to eachother or as both bad. We exclude all such responses from Pure Dove and make sure to only include\u2026 See the full description on the dataset page: https://huggingface.co/datasets/LDJnr/Pure-Dove."
+                }
+            ]
+        },
+        {
+            "type": "data",
+            "bom-ref": "LDJnr/Verified-Camel-6c954918-6150-534c-952b-ea2cff17c0b2",
+            "name": "LDJnr/Verified-Camel",
+            "data": [
+                {
+                    "type": "dataset",
+                    "bom-ref": "LDJnr/Verified-Camel-6c954918-6150-534c-952b-ea2cff17c0b2",
+                    "name": "LDJnr/Verified-Camel",
+                    "contents": {
+                        "url": "https://huggingface.co/datasets/LDJnr/Verified-Camel",
+                        "properties": [
+                            {
+                                "name": "task_categories",
+                                "value": "conversational, question-answering, text-generation"
+                            },
+                            {
+                                "name": "language",
+                                "value": "en"
+                            },
+                            {
+                                "name": "size_categories",
+                                "value": "n<1K"
+                            },
+                            {
+                                "name": "pretty_name",
+                                "value": "Verified-Camel"
+                            },
+                            {
+                                "name": "license",
+                                "value": "apache-2.0"
+                            }
+                        ]
+                    },
+                    "governance": {
+                        "owners": [
+                            {
+                                "organization": {
+                                    "name": "LDJnr",
+                                    "url": "https://huggingface.co/LDJnr"
+                                }
+                            }
+                        ]
+                    },
+                    "description": "\n\t\n\t\t\n\t\tThis is the Official Verified Camel dataset. Just over 100 verified examples, and many more coming soon!\n\t\n\n\nComprised of over 100 highly filtered and curated examples from specific portions of CamelAI stem datasets. \n\nThese examples are verified to be true by experts in the specific related field, with atleast a bachelors degree in the subject.\n\nRoughly 30-40% of the originally curated data from CamelAI was found to have atleast minor errors and/or incoherent questions(as determined\u2026 See the full description on the dataset page: https://huggingface.co/datasets/LDJnr/Verified-Camel."
+                }
+            ]
+        },
+        {
+            "type": "data",
+            "bom-ref": "tiedong/goat-38642401-79d7-541d-a92f-d3d7acdb8db8",
+            "name": "tiedong/goat",
+            "data": [
+                {
+                    "type": "dataset",
+                    "bom-ref": "tiedong/goat-38642401-79d7-541d-a92f-d3d7acdb8db8",
+                    "name": "tiedong/goat",
+                    "contents": {
+                        "url": "https://huggingface.co/datasets/tiedong/goat",
+                        "properties": [
+                            {
+                                "name": "task_categories",
+                                "value": "question-answering"
+                            },
+                            {
+                                "name": "language",
+                                "value": "en"
+                            },
+                            {
+                                "name": "size_categories",
+                                "value": "1M<n<10M"
+                            },
+                            {
+                                "name": "license",
+                                "value": "apache-2.0"
+                            }
+                        ]
+                    },
+                    "governance": {
+                        "owners": [
+                            {
+                                "organization": {
+                                    "name": "tiedong",
+                                    "url": "https://huggingface.co/tiedong"
+                                }
+                            }
+                        ]
+                    },
+                    "description": "\n\t\n\t\t\n\t\tDataset Card for Dataset Name\n\t\n\n\n\t\n\t\t\n\t\tDataset Summary\n\t\n\nThe dataset.json file contains ~1.7 million synthetic data for arithmetic tasks, generated by dataset.ipynb.\n\n\t\n\t\t\n\t\tSupported Tasks and Leaderboards\n\t\n\n[More Information Needed]\n\n\t\n\t\t\n\t\tLanguages\n\t\n\n[More Information Needed]\n\n\t\n\t\t\n\t\tDataset Structure\n\t\n\n\n\t\n\t\t\n\t\tData Instances\n\t\n\n[More Information Needed]\n\n\t\n\t\t\n\t\tData Fields\n\t\n\n[More Information Needed]\n\n\t\n\t\t\n\t\tData Splits\n\t\n\n[More Information Needed]\n\n\t\n\t\t\n\t\tDataset Creation\u2026 See the full description on the dataset page: https://huggingface.co/datasets/tiedong/goat."
+                }
+            ]
+        },
+        {
+            "type": "data",
+            "bom-ref": "glaiveai/glaive-code-assistant-4cf9eb44-c4b0-5c31-b48c-ce95c2926a6d",
+            "name": "glaiveai/glaive-code-assistant",
+            "data": [
+                {
+                    "type": "dataset",
+                    "bom-ref": "glaiveai/glaive-code-assistant-4cf9eb44-c4b0-5c31-b48c-ce95c2926a6d",
+                    "name": "glaiveai/glaive-code-assistant",
+                    "contents": {
+                        "url": "https://huggingface.co/datasets/glaiveai/glaive-code-assistant",
+                        "properties": [
+                            {
+                                "name": "size_categories",
+                                "value": "100K<n<1M"
+                            },
+                            {
+                                "name": "license",
+                                "value": "apache-2.0"
+                            }
+                        ]
+                    },
+                    "governance": {
+                        "owners": [
+                            {
+                                "organization": {
+                                    "name": "glaiveai",
+                                    "url": "https://huggingface.co/glaiveai"
+                                }
+                            }
+                        ]
+                    },
+                    "description": "\n\t\n\t\t\n\t\tGlaive-code-assistant\n\t\n\nGlaive-code-assistant is a dataset of ~140k code problems and solutions generated using Glaive\u2019s synthetic data generation platform.\nThe data is intended to be used to make models act as code assistants, and so the data is structured in a QA format where the questions are worded similar to how real users will ask code related questions.\nThe data has ~60% python samples.\nTo report any problems or suggestions in the data, join the Glaive discord\n"
+                }
+            ]
+        },
+        {
+            "type": "data",
+            "bom-ref": "meta-math/MetaMathQA-c6cf810a-8b06-5552-a876-53681c5fe9a1",
+            "name": "meta-math/MetaMathQA",
+            "data": [
+                {
+                    "type": "dataset",
+                    "bom-ref": "meta-math/MetaMathQA-c6cf810a-8b06-5552-a876-53681c5fe9a1",
+                    "name": "meta-math/MetaMathQA",
+                    "contents": {
+                        "url": "https://huggingface.co/datasets/meta-math/MetaMathQA",
+                        "properties": [
+                            {
+                                "name": "license",
+                                "value": "mit"
+                            }
+                        ]
+                    },
+                    "governance": {
+                        "owners": [
+                            {
+                                "organization": {
+                                    "name": "meta-math",
+                                    "url": "https://huggingface.co/meta-math"
+                                }
+                            }
+                        ]
+                    },
+                    "description": "View the project page:\nhttps://meta-math.github.io/\nsee our paper at https://arxiv.org/abs/2309.12284\n\n\t\n\t\t\n\t\tNote\n\t\n\nAll MetaMathQA data are augmented from the training sets of GSM8K and MATH. \nNone of the augmented data is from the testing set.\nYou can check the original_question in meta-math/MetaMathQA, each item is from the GSM8K or MATH train set.\n\n\t\n\t\t\n\t\tModel Details\n\t\n\nMetaMath-Mistral-7B is fully fine-tuned on the MetaMathQA datasets and based on the powerful Mistral-7B model. It is\u2026 See the full description on the dataset page: https://huggingface.co/datasets/meta-math/MetaMathQA."
+                }
+            ]
+        },
+        {
+            "type": "data",
+            "bom-ref": "OpenAssistant/oasst_top1_2023-08-25-10cff0b1-2289-5b35-8ea8-59253cea318a",
+            "name": "OpenAssistant/oasst_top1_2023-08-25",
+            "data": [
+                {
+                    "type": "dataset",
+                    "bom-ref": "OpenAssistant/oasst_top1_2023-08-25-10cff0b1-2289-5b35-8ea8-59253cea318a",
+                    "name": "OpenAssistant/oasst_top1_2023-08-25",
+                    "contents": {
+                        "url": "https://huggingface.co/datasets/OpenAssistant/oasst_top1_2023-08-25",
+                        "properties": [
+                            {
+                                "name": "task_categories",
+                                "value": "conversational"
+                            },
+                            {
+                                "name": "size_categories",
+                                "value": "10K<n<100K"
+                            },
+                            {
+                                "name": "license",
+                                "value": "apache-2.0"
+                            }
+                        ]
+                    },
+                    "governance": {
+                        "owners": [
+                            {
+                                "organization": {
+                                    "name": "OpenAssistant",
+                                    "url": "https://huggingface.co/OpenAssistant"
+                                }
+                            }
+                        ]
+                    },
+                    "description": "\n\t\n\t\t\n\t\tOpenAssistant TOP-1 Conversation Threads\n\t\n\n\nGuanacco style export of the best conversation threads from the open-assistant.io database\nexported August 25, 2023\njsonl files with chatml formatted conversations\ntrain: 12,947 samples / valid: 680 samples\n\n"
+                }
+            ]
+        },
+        {
+            "type": "data",
+            "bom-ref": "TIGER-Lab/MathInstruct-9d9c997d-f6c1-5029-96fd-6003c4f0ec06",
+            "name": "TIGER-Lab/MathInstruct",
+            "data": [
+                {
+                    "type": "dataset",
+                    "bom-ref": "TIGER-Lab/MathInstruct-9d9c997d-f6c1-5029-96fd-6003c4f0ec06",
+                    "name": "TIGER-Lab/MathInstruct",
+                    "contents": {
+                        "url": "https://huggingface.co/datasets/TIGER-Lab/MathInstruct",
+                        "properties": [
+                            {
+                                "name": "task_categories",
+                                "value": "text-generation"
+                            },
+                            {
+                                "name": "language",
+                                "value": "en"
+                            },
+                            {
+                                "name": "size_categories",
+                                "value": "100K<n<1M"
+                            },
+                            {
+                                "name": "pretty_name",
+                                "value": "MathInstruct"
+                            },
+                            {
+                                "name": "license",
+                                "value": "mit"
+                            }
+                        ]
+                    },
+                    "governance": {
+                        "owners": [
+                            {
+                                "organization": {
+                                    "name": "TIGER-Lab",
+                                    "url": "https://huggingface.co/TIGER-Lab"
+                                }
+                            }
+                        ]
+                    },
+                    "description": "\n\t\n\t\t\n\t\t\ud83e\udda3 MAmmoTH: Building Math Generalist Models through Hybrid Instruction Tuning\n\t\n\nMathInstruct is a meticulously curated instruction tuning dataset that is lightweight yet generalizable. MathInstruct is compiled from 13 math rationale datasets, six of which are newly curated by this work. It uniquely focuses on the hybrid use of chain-of-thought (CoT) and program-of-thought (PoT) rationales, and ensures extensive coverage of diverse mathematical fields. \nProject Page:\u2026 See the full description on the dataset page: https://huggingface.co/datasets/TIGER-Lab/MathInstruct."
+                }
+            ]
+        }
+    ]
+}