License

The model is licensed under the MIT license.

Intended Uses

Primary use cases The model is intended for broad commercial and research use in English. The model provides uses for general purpose AI systems and applications which require:

1. Memory/compute constrained environments
2. Latency bound scenarios
3. Strong reasoning (especially code, math and logic)

Our model is designed to accelerate research on language and multimodal models, for use as a building block for generative AI powered features. Out-of-scope use cases Our models are not specifically designed or evaluated for all downstream purposes. Developers should consider common limitations of language models as they select use cases, and evaluate and mitigate for accuracy, safety, and fairness before using within a specific downstream use case, particularly for high-risk scenarios. Developers should be aware of and adhere to applicable laws or regulations (including privacy, trade compliance laws, etc.) that are relevant to their use case. Nothing contained in this Model Card should be interpreted as or deemed a restriction or modification to the license the model is released under.

Trademarks

This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines . Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party's policies.

Responsible AI Considerations

Like other language models, the Phi series models can potentially behave in ways that are unfair, unreliable, or offensive. Some of the limiting behaviors to be aware of include:

• Quality of Service: the Phi models are trained primarily on English text. Languages other than English will experience worse performance. English language varieties with less representation in the training data might experience worse performance than standard American English.
• Representation of Harms & Perpetuation of Stereotypes: These models can over- or under-represent groups of people, erase representation of some groups, or reinforce demeaning or negative stereotypes. Despite safety post-training, these limitations may still be present due to differing levels of representation of different groups or prevalence of examples of negative stereotypes in training data that reflect real-world patterns and societal biases.
• Inappropriate or Offensive Content: these models may produce other types of inappropriate or offensive content, which may make it inappropriate to deploy for sensitive contexts without additional mitigations that are specific to the use case.
• Information Reliability: Language models can generate nonsensical content or fabricate content that might sound reasonable but is inaccurate or outdated.
• Limited Scope for Code: Majority of Phi-3 training data is based in Python and use common packages such as "typing, math, random, collections, datetime, itertools". If the model generates Python scripts that utilize other packages or scripts in other languages, we strongly recommend users manually verify all API uses.

Developers should apply responsible AI best practices and are responsible for ensuring that a specific use case complies with relevant laws and regulations (e.g. privacy, trade, etc.). Important areas for consideration include:

• Allocation: Models may not be suitable for scenarios that could have consequential impact on legal status or the allocation of resources or life opportunities (ex: housing, employment, credit, etc.) without further assessments and additional debiasing techniques.
• High-Risk Scenarios: Developers should assess suitability of using models in high-risk scenarios where unfair, unreliable or offensive outputs might be extremely costly or lead to harm. This includes providing advice in sensitive or expert domains where accuracy and reliability are critical (ex: legal or health advice). Additional safeguards should be implemented at the application level according to the deployment context.
• Misinformation: Models may produce inaccurate information. Developers should follow transparency best practices and inform end-users they are interacting with an AI system. At the application level, developers can build feedback mechanisms and pipelines to ground responses in use-case specific, contextual information, a technique known as Retrieval Augmented Generation (RAG).
• Generation of Harmful Content: Developers should assess outputs for their context and use available safety classifiers or custom solutions appropriate for their use case.
• Misuse: Other forms of misuse such as fraud, spam, or malware production may be possible, and developers should ensure that their applications do not violate applicable laws and regulations.

Content Filtering

Prompts and completions are passed through a default configuration of Azure AI Content Safety classification models to detect and prevent the output of harmful content. Learn more about Azure AI Content Safety . Configuration options for content filtering vary when you deploy a model for production in Azure AI; learn more .

We report the results under completion format for Phi-3-Mini-128K-Instruct on standard open-source benchmarks measuring the model's reasoning ability (both common sense reasoning and logical reasoning). We compare to Mistral-7b-v0.1, Mixtral-8x7b, Gemma 7B, Llama-3-8B-Instruct, and GPT-3.5. All the reported numbers are produced with the exact same pipeline to ensure that the numbers are comparable. These numbers might differ from other published numbers due to slightly different choices in the evaluation. As is now standard, we use few-shot prompts to evaluate the models, at temperature 0.
The prompts and number of shots are part of a Microsoft internal tool to evaluate language models, and in particular we did no optimization to the pipeline for Phi-3.
More specifically, we do not change prompts, pick different few-shot examples, change prompt format, or do any other form of optimization for the model. The number of k–shot examples is listed per-benchmark.

Category	Benchmark	Phi-3-Mini-128K-Ins	Gemma-7B	Mistral-7B	Mixtral-8x7B	Llama-3-8B-Ins	GPT3.5-Turbo-1106
Popular aggregated benchmark	AGI Eval 5-shot	39.5	42.1	35.1	45.2	42	48.4
	MMLU 5-shot	69.7	63.6	61.7	70.5	66.5	71.4
	BigBench Hard 3-shot	72.1	59.6	57.3	69.7	51.5	68.3
Language Understanding	ANLI 7-shot	52.3	48.7	47.1	55.2	57.3	58.1
	HellaSwag 5-shot	70.5	49.8	58.5	70.4	71.1	78.8
Reasoning	ARC Challenge 10-shot	85.5	78.3	78.6	87.3	82.8	87.4
	BoolQ 0-shot	77.1	66	72.2	76.6	80.9	79.1
	MedQA 2-shot	56.4	49.6	50	62.2	60.5	63.4
	OpenBookQA 10-shot	78.8	78.6	79.8	85.8	82.6	86
	PIQA 5-shot	80.1	78.1	77.7	86	75.7	86.6
	GPQA 0-shot	29.7	2.9	15	6.9	32.4	29.9
	Social IQA 5-shot	74.7	65.5	74.6	75.9	73.9	68.3
	TruthfulQA (MC2) 10-shot	64.8	52.1	53	60.1	63.2	67.7
	WinoGrande 5-shot	71.0	55.6	54.2	62	65	68.8
Factual Knowledge	TriviaQA 5-shot	57.8	72.3	75.2	82.2	67.7	85.8
Math	GSM8K CoTT 8-shot	85.3	59.8	46.4	64.7	77.4	78.1
Code Generation	HumanEval 0-shot	60.4	34.1	28.0	37.8	60.4	62.2
	MBPP 3-shot	70.0	51.5	50.8	60.2	67.7	77.8
Average		66.4	56.0	56.4	64.4	65.5	70.3

Long Context: Phi-3 Mini-128K-Instruct supports 128K context length, therefore the model is capable of several long context tasks including long document/meeting summarization, long document QA.

Benchmark	Phi-3 Mini-128K-Instruct	Mistral-7B	Mixtral 8x7B	LLaMA-3-8B-Instruct
GovReport	25.3	4.9	20.3	10.3
QMSum	21.9	15.5	20.6	2.9
Qasper	41.6	23.5	26.6	8.1
SQuALITY	24.1	14.7	16.2	25
SummScreenFD	16.8	9.3	11.3	5.1
Average	25.9	13.6	19.0	10.3

We take a closer look at different categories across 100 public benchmark datasets at the table below:

Category	Phi-3-Mini-128K-Instruct	Gemma-7B	Mistral-7B	Mixtral 8x7B	Llama-3-8B-Instruct	GPT-3.5-Turbo
Popular aggregated benchmark	60.6	59.4	56.5	66.2	59.9	67.0
Reasoning	69.4	60.3	62.8	68.1	69.6	71.7
Language understanding	57.5	57.6	52.5	66.1	63.2	67.7
Code generation	61.0	45.6	42.9	52.7	56.4	70.4
Math	51.6	35.8	25.4	40.3	41.1	52.8
Factual knowledge	35.8	46.7	49.8	58.6	43.1	63.4
Multilingual	56.4	66.5	57.4	66.7	66.6	71.0
Robustness	61.1	38.4	40.6	51.0	64.5	69.3

Overall, the model with only 3.8B-param achieves a similar level of language understanding and reasoning ability as much larger models. However, it is still fundamentally limited by its size for certain tasks. The model simply does not have the capacity to store too much world knowledge, which can be seen for example with low performance on TriviaQA. However, we believe such weakness can be resolved by augmenting Phi-3-Mini with a search engine.