Public AI Models - AI Tools For Small Business

Public AI LLM models:

best use cases,
specialised tasks,
tips for the best outputs.

1. Cohere Summarize

- Best use cases: Text summarisation, creating concise summaries of longer texts.
- Specialized tasks: Summarizing documents, articles, or other text input.
- Tips for best output: Provide clear instructions and specify the desired length of the summary. Make sure the input is well-structured and coherent.

2. GPT3.5-Turbo

- Best use cases: Natural language understanding, text completion, question-answering, conversational agents.
- Specialised tasks: Generating human-like responses, language translation, and chatbot interactions.
- Tips for best output: Clearly define your prompt or question, provide context when necessary, and consider experimenting with different temperature settings for generating responses.

3. GPT2

- Best use cases: Language generation, creative writing, content creation.
- Specialised tasks: Generating text, completing sentences, creating stories.
- Tips for best output: Start with a clear and concise prompt, experiment with different temperature settings to control the randomness of generated text, and consider adjusting the length of the generated output.

4. GGPT2 Large

- Best use cases: Language generation, text completion, dialogue generation.
- Specialised tasks: Generating more complex and detailed text, and interactive conversations.
- Tips for best output: Provide specific instructions and context, experiment with different temperature settings, and consider adjusting the output length.

5. GPT2 XL

- Best use cases: Language generation, text completion, dialogue generation.
- Specialised tasks: Generating longer, more coherent text and complex story generation.
- Tips for best output: Clearly define the desired output, provide detailed instructions and context, and experiment with temperature settings and length adjustment.

6. Pegasus paraphrase

- Best use cases: Paraphrasing, rephrasing sentences or text.
- Specialised tasks: Generating alternative versions of sentences or texts while retaining the original meaning.
- Tips for best output: Specify the input sentence or text, provide context if necessary, and experiment with different temperature settings.

7. OpenAI GPT

- Best use cases: Language generation, text completion, creative writing.
- Specialised tasks: Generating coherent and context-aware text.
- Tips for best output: Clearly define the prompt, provide context if needed, and experiment with different temperature settings to control the output's creativity.

8. Bart Large CNN

- Best use cases: Language generation, abstractive summarisation.
- Specialised tasks: Generating summaries of documents, articles, or text inputs.
- Tips for best output: Specify the input text or document, provide summarisation instructions, and experiment with different settings for length and coherence.

9. DialoGPT Medium

- Best use cases: Conversational agents, chatbot interactions, dialogue generation.
- Specialised tasks: Generating responses in conversational contexts, interactive conversations.
- Tips for best output: Clearly define the conversation history and the desired response, provide context, and experiment with different temperature settings for generating responses.

10. Roberta Base Squad2

- Best use cases: Question-answering, reading comprehension.
- Specialised tasks: Answering questions based on given context or passages.
- Tips for best output: Clearly specify the question and provide relevant context or passages for answering, experiment with different temperature settings, and consider truncating the input if it exceeds the model's maximum token limit.

11. Cohere Base Light:

12. Llama

- Best use cases: Language generation, text completion, conversation modeling.
- Specialised tasks: Generating natural and context-aware responses to chatbot interactions.
- Tips for best output: Provide specific instructions and context, experiment with different temperature settings, and consider adjusting the output length.

13. Claude2

- Best use cases: Language generation, creative writing, dialogue generation.
- Specialised tasks: Generating text focusing on creativity, interactive conversations.
- Tips for best output: Clearly define the desired output, provide detailed instructions and context, and experiment with temperature settings and length adjustment.

14. Google Bard

- Best use cases: Language generation, dialogue generation, coding support, image upload, and internet access.
- Specialised tasks: Writing poems, solving puzzles, providing travel advice, and assisting with spreadsheet formulas.
- Tips for best output: Provide as much information as possible in the prompt, add specific details for more relevant output, and always double-check the sources of information provided by AI

Please note that these summaries are based on general knowledge of the models. For optimal usage, it is recommended to consult the official documentation and guidelines provided by the model developers.

IMAGE GENERATORS

1. DALL-E:

- Best use cases: Image generation, creating unique and surreal images from textual prompts.
- Specialised tasks: Transforming textual descriptions into corresponding images.
- Tips for best output: Provide clear and specific textual prompts describing the desired image, experiment with different prompts and parameters, and be aware that generating high-resolution images may require more computational resources.

2. Stable Diffusion 1.5

- Best use cases: Image editing, enhancing image quality, removing noise, and restoring images.
- Specialized tasks: Image denoising, inpainting, super-resolution, and style transfer.
- Tips for best output: Clearly define the editing task, provide relevant inputs or constraints, experiment with different parameters, and be prepared for longer computation times for complex tasks.

3. Midjourney

- Best use cases: Image classification, object recognition, and visual understanding.
- Specialised tasks: Analyzing and categorising images based on their content.
- Tips for best output: Provide clear and well-defined images for classification, ensure images are of good quality and resolution, and consider experimenting with different preprocessing techniques for optimal results.

Please note that these summaries are based on general knowledge of the models, and it is advisable to refer to the official documentation and guidelines provided by the model developers for specific usage instructions and best practices.

Public AI LLM & Image Generation Models