What are the hardware requirements for Sum Small?

Sum Small requires approximately 7.7 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Sum Small and how large is it?

Sum Small is developed by omi-health, a model with 3.8b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Sum Small?

Sum Small is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Sum Small by omi-health — VRAM 7.7GB, 4K context

Name: Sum Small
Rating: 4 (3 reviews)
Author: omi-health

Sum Small is an open-source language model by omi-health. Features: 3.8b LLM, VRAM: 7.7GB, Context: 4K, License: mit, Instruction-Based, LLM Explorer Score: 0.12.

Conversational Custom code Dataset:omi-health/medical-dia... En Endpoints compatible Instruct Phi3 Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/omi-health/sum-small

Sum Small Parameters and Internals

Model Type

text generation

Use Cases

Areas:

research, AI-powered medical documentation

Primary Use Cases:

generation of SOAP summaries from medical dialogues

Limitations:

Synthetic data, not for direct clinical use

Considerations:

Integration with additional safety guardrails needed.

Supported Languages

en (full-proficiency)

Training Details

Data Sources:

Omi Health/medical-dialogue-to-soap-summary

Data Volume:

10,000 dialogues

Methodology:

fine-tuning

Hardware Used:

NVIDIA A100 GPUs

Responsible Ai Considerations

Fairness:

Users are urged to consider ethical implications in healthcare use.

Transparency:

Model's synthetic training data composition.

Accountability:

Not directly accountable without safety validation.

LLM Name	Sum Small
Repository 🤗	https://huggingface.co/omi-health/sum-small
Model Size	3.8b
Required VRAM	7.7 GB
Updated	2026-08-06
Maintainer	omi-health
Model Type	phi3
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-2 2.7 GB: 2-of-2
Supported Languages	en
Model Architecture	Phi3ForCausalLM
License	mit
Context Length	4096
Model Max Length	4096
Transformers Version	4.41.0.dev0
Tokenizer Class	LlamaTokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	32064
Torch Data Type	bfloat16

Rank the Sum Small capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Sum Small

Best Alternatives	Context / RAM	Downloads	Likes
Phi 4 Mini Instruct	128K / 7.7 GB	514603	791
Phi 3.5 Mini Instruct	128K / 7.7 GB	1164079	999
Phi 3 Mini 128K Instruct	128K / 7.7 GB	233663	1700
Phi 3 Mini 128K Instruct	128K / 7.7 GB	7	0
Phi 4 Mini Abliterated	128K / 7.7 GB	33	1
Phi 3 Mini 128K Instruct	128K / 7.7 GB	11	0
Phi 4 Mini Instruct Heretic	128K / 7.7 GB	13	0
Phi 3.5 Mini TitanFusion 0.1	128K / 7.7 GB	7	0
NuExtract V1.5	128K / 7.7 GB	108511	89
MediPhi Instruct	128K / 7.7 GB	1671	69

Note: green Score (e.g. "73.2") means that the model is better than omi-health/sum-small.