Mirrors/unleash.unleash
1
0
Fork 0
mirror of https://github.com/Unleash/unleash.git synced 2025-09-19 17:52:45 +02:00
Code Issues Projects Releases Wiki Activity

Copy edits

Browse Source
This commit is contained in:
Alvin Bryan 2024-10-16 19:22:02 +01:00
parent 957254bc63
commit ffaf42eb5e
No known key found for this signature in database
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
website/docs/feature-flag-tutorials/use-cases/ai.md
View File

@ -5,12 +5,12 @@ slug: /feature-flag-tutorials/use-cases/ai
Many tech companies today are integrating generative AI with large language models (LLMs) into their products, often in the form of chatbots or content generators. Many tech companies today are integrating generative AI with large language models (LLMs) into their products, often in the form of chatbots or content generators.
The main way to interact with LLMs today is via a set of APIs, usually either OpenAI, Anthropic or aggregators like Groq. All the APIs usually have similar parameters, like: The main way to interact with LLMs today is via a set of APIs, usually either OpenAI, Anthropic or aggregators like Groq. Most of these APIs have similar parameters, like:
- `model`: The model and the specific version of the model to use. - `model`: The model and the specific version of the model to use.
- `prompt`: The prompt to give to the LLM. - `prompt`: The prompt to give to the LLM.
Because all these models vary a lot in their capabilities, chances are you'll be testing multiple models, and multiple versions of the same model, and multiple prompts. Given how much these models vary in their capabilities, chances are you'll be testing multiple models, multiple versions of the same model, and multiple prompts.
This is where feature flags are super useful. They allow you to easily switch between different configurations. In this tutorial, we’ll explore how to use feature flags with AI models. This is where feature flags are super useful. They allow you to easily switch between different configurations. In this tutorial, we’ll explore how to use feature flags with AI models.
@ -98,7 +98,7 @@ When a request comes in, it:
- Uses the OpenAI API to generate a response. - Uses the OpenAI API to generate a response.
- Streams the response back to the client. - Streams the response back to the client.
The `streamText` function part of some utilities provided by Vercel's AI SDK, which helps deal with real-time streaming of the AI's responses and other AI-related tasks. The `streamText` function is one of the utilities provided by Vercel's AI SDK. It helps deal with the real-time streaming of the AI's responses.
## Creating a feature flag with AI variants ## Creating a feature flag with AI variants