New Large Language Models (LLMs)

The Talkative platform can now integrate with multiple Large Language Models (LLMs).

Previously, we only supported OpenAI’s GPT-3.5 and GPT-4 models.

Now, we’ve added OpenAI’s most recent release: GPT-4o.

In addition to multiple GPT options, we also offer AI models from Anthropic (Claude variants) and Meta (Llama variants).

It means you’ll have ultimate flexibility to tailor our AI solutions to your specific use cases. 

‍

Choosing the best LLM for your needs

For various Talkative features (e.g. AI Agent Copilot, GenAI Chatbots), you can now choose between the following AI models...

• GPT-4
• GPT-4o
• GPT-4o Mini
• LLAMA-3 (70B)
• LLAMA-3 (80B)
• Claude Haiku
• Claude Sonnet
• Google Gemini 1.5 Flash
• Google Gemini 1.5 Flo

Which LLM is best depends on your requirements and priorities.

Generally speaking, Claude Haiku, and Llama-3.8 are the most cost-effective options.

However, although powerful, these models are not as advanced as the other LLMs when it comes to reasoning and precise instructions.

This means that GPT-4, GPT-4o, Llama-3.70, or Claude Sonnet may be more suitable for particularly complex use cases.

Gemini 1.5 Flash and Gemini 1.5 Flo are similar in their capabilities to the GPT models, except they can also be regionalised.

The best approach is to weigh up the pros and cons of each model against your needs and budget.

Getting started 

Not sure which LLM is right for you?

Get in touch with your Talkative Account Manager for advice and guidance.

And for more on our latest updates, check out our most recent release notes.

Are you a new customer who wants to learn more? 

Book a demo today, and check out our interactive product tour.