Use and fine-tune small, task-specific LLMs. Without training data or GPUs.

Reduce LLM API costs, get built-in privacy and improve performance on your tasks by using task-specific, local, small LLMs.

Use pre-trained, task-specific small LLMs on your CPU for fast and cost-effective inference.

Fine-tune the pre-trained models without training data; models are trained on synthetic data generated on-the-fly.

Stop sending data to third-party servers. Since models run locally on your CPU, your data never leaves your machine.

Offload tasks to local, specialized task-specific models; reduce cost on your LLM API, while improving accuracy.

Task	Model	Size	Description	Inference on CPU	Fine-Tune on CPU
Text Classification	No default model — must be trained	0.1B params, 470Mb	Performs general-purpose text classification based on the user requirements.	Code Examples	Code Examples
Guardrail	tanaos/tanaos-guardrail-v1	0.1B params, 500Mb	Flags unsafe, harmful, or off-topic messages.	Code Examples	Code Examples
Intent Classification	tanaos/tanaos-intent-classifier-v1	0.1B params, 500Mb	Classifies user messages into predefined intent categories.	Code Examples	Code Examples
Reranker	cross-encoder/mmarco-mMiniLMv2-L12-H384-v1	0.1B params, 470Mb	Ranks a list of items or search results based on relevance to a query.	Code Examples	Code Examples
Sentiment Analysis	tanaos/tanaos-sentiment-analysis-v1	0.1B params, 470Mb	Determines the sentiment (positive, negative, neutral) of a given text.	Code Examples	Code Examples
Emotion Detection	tanaos/tanaos-emotion-detection-v1	0.1B params, 470Mb	Identifies the emotion expressed in a given text.	Code Examples	Code Examples
Named Entity Recognition	tanaos/tanaos-NER-v1	0.1B params, 500Mb	Detects and classifies named entities in text.	Code Examples	Code Examples
Text Anonymization	tanaos/tanaos-text-anonymizer-v1	0.1B params, 500Mb	Removes personally identifiable information (PII) from text.	Code Examples	Code Examples

Install Artifex with pip install artifex and use it for free. Read the docs to learn how to integrate it into your project.

Get started by creating an account on our platform.

Pick a plan to get access to higher rate limits and better model performance.

Create an api key and pass it to Artifex at initialization.