
Generative AI with Cohere: Part 1 - Model Prompting
In this multi-part guide, we will go through everything that you need to know about generative AI with Cohere’s large language models (LLMs). In Part 1, we talk about model prompting.
What is Generative AI?
Generative AI is a type of artificial intelligence that focuses on creating or generating new content or data. This can be in the form of language, images, videos, and more.
Its market potential is significant as it has the potential to revolutionize many industries and drive innovation in a wide range of fields. For example, in creative arts, generative AI can be used to generate unique and engaging content, such as music or visual art, with minimal need for human input. In the business world, generative AI can be used to generate reports, presentations, and other business documents, reducing the need for manual data analysis and enhancing productivity.
At Cohere, our focus is on language. We want to enable developers to add language AI to their technology stack and build impactful applications with it.
In this multi-part guide, we will go through everything that you need to know about generative AI with Cohere’s large language models (LLMs).
Here is what we’ll cover throughout this series:
- Part 1: Model Prompting
- Part 2: Use Case Ideation
- Part 3: The Generate Endpoint
- Part 4: Creating Custom Models
- Part 5: Chaining Prompts

In this Part 1 article, we will cover the following topics:
- Getting Started with the Cohere Playground
- Prompting the Models
- Controlling the Model Output
- Saving and Sharing Your Prompts
Getting Started with the Cohere Playground
Throughout the series, we will cover the full spectrum of working with generative AI to enable you to build applications with it. But to start with, let’s take the no-code route: we’ll show you how AI text generation works, and how you can experiment with it in the Cohere Playground.
First, sign up for a Cohere account and then visit the Playground.
The Playground UI consists of a few sections. The main window is where you enter your prompt and where the output, or response, is generated. A menu of saved prompts, or presets
, is shown above the prompt window, and model parameter controls are located in the right-hand pane.

We’ll cover presets and parameters later in this article, but now, let’s start with the fun part: prompt design.
Prompting the Models
Prompt Design
Prompting is at the heart of working with LLMs. The prompt provides a context for the text that we want the model to generate. The prompts we create can be anything from simple instructions to more complex pieces of text, and they are used to encourage the model to produce a specific type of output.
Coming up with a good prompt is a bit of both science and art. On the one hand, we know the broad patterns that enable us to construct a prompt that will generate the output that we want. But on the other hand, there is so much room for creativity and imagination in coming up with prompts that can get the best out of a model.
With that, let’s now dive into how you can design more effective prompts.
Prompting the Command model by Instruction
The best way to design prompts is to tell a model like Command
what you need it to do like you’re giving the model a command or an instruction. One way to do this is by using imperative verbs to tell the model what to do, for example: generate, write, list, provide, and other variations.

Let’s say that we are creating the product description copy for a wireless earbuds product. We can write the prompt as follows.
Generate a concise product description for the product: wireless earbuds.
At this point, ensure that you select command
in the MODEL
dropdown on the right pane. Then, click on Run.
This generates the following output.

That’s not bad. With a simple, one-line prompt, we already have a piece of product description that can make a digital marketer proud!
But perhaps we want to be more specific in terms of what we want the output to look like. For this, we can layer additional instructions onto the model in the prompt.
Let’s say that we want the model to write the product description in a certain format with specific information. In this case, we can append this specific instruction in the prompt as follows.
Generate a concise product description for the product: wireless earbuds. Use the following format: Hook, Solution, Features and Benefits, Call to Action.

The model duly returns an output following the format that we wanted.
The prompt can also be constructed as a combination of an instruction and some context. Let’s see this in action with another example: emails. We can create a simple instruction to write an email as follows.

Or we can create another prompt to summarize that email, which now becomes part of the prompt, acting as the context.

This instruction–context prompt format is extremely useful as it means we can insert additional information into the context to help ground the model's output. One such example is a question-answering system for, let's say, a company's knowledge base. Given a question (the instruction), the model will only be able to provide accurate answers if provided with the knowledge base (the context).
Let's move to another example—an extraction task, which is another type of task that a generative model can do very well. Given a context, which in this case is a description of a movie, we want the model to extract the movie title. Here's the prompt and its output.
Extract the movie title from the text below. Deadpool 2 | Official HD Deadpool's "Wet on Wet" Teaser | 2018

The model is also effective at tasks that involve taking a piece of text and rewriting it into another format that we need.
Here's an example. We have a one-line instruction followed by the context, which in this case is a blog excerpt. The instruction is to generate a list of frequently asked questions (FAQ) based on the passage, which involves a mixture of several tasks such as extraction and rewriting. Here's the prompt.
Given the following text, write down a list of potential frequently asked questions (FAQ), together with the answers.
The Cohere Platform provides an API for developers and organizations to access cutting-edge LLMs without needing machine learning know-how. The platform handles all the complexities of curating massive amounts of text data, model development, distributed training, model serving, and more. This means that developers can focus on creating value on the applied side rather than spending time and effort on the capability-building side.
There are two key types of language processing capabilities that the Cohere Platform provides — text generation and text embedding — and each is served by a different type of model.
With text generation, we enter a piece of text, or prompt, and get back a stream of text as a completion to the prompt. One example is asking the model to write a haiku (the prompt) and getting an originally written haiku in return (the completion).
With text embedding, we enter a piece of text and get back a list of numbers that represents its semantic meaning (we’ll see what “semantic” means in a section below). This is useful for use cases that involve “measuring” what a passage of text represents, for example, in analyzing its sentiment.
And the result is a nice list of FAQs about the blog excerpt.

By now we can see how versatile model is at performing various forms of tasks—not just freeform text generation, but also following instructions, working with contextual information, summarizing long passages, extracting information, rewriting text into different forms, and more.
This is just a taste of what kinds of prompts you can design. You can keep layering your instructions to be as specific as you want, and see the output generated by the model. And there is really no right or wrong way to design a prompt. It’s really about applying an idea and continuing to iterate the prompt until you get the outcome you are looking for.
Prompting by Example
For historical context, it’s important to point out that before instruction-tuned models were released, earlier language models were prompted in a different style that relied on including examples of the desired output in the prompt.
For those earlier models,base
and base-light
, a basic prompt format that generally works well is as follows.
A short description of the overall context
- A few examples of prompts and completions; usually two to three examples are sufficient but for more challenging tasks, you will need more
- A short sequence of characters or “stop sequences” to guide the model towards writing a complete passage and then stopping.
Controlling the Model Output
Other than the prompt design, there is another way to control the kind of output we want, that is, by adjusting the model parameters. These parameters are available on the right pane of the Playground. They are applicable to both types of prompting we discussed earlier.
Let’s now see what we can do with these parameters.
When to Stop
There are a couple of parameters that let you decide when the model should stop.
NUMBER OF WORDS
— The model will stop generating text once it reaches the maximum number of words specified by this parameter. Note: the model treats words astokens
. One word is about three tokens.STOP SEQUENCE
— You can define any character or sequence of characters to tell the model when to stop. This is useful when you are showing the model a few generation examples and you want to show exactly where one example ends. For examples, you can use “--” to split the examples, or simply a new line (\n
character) or a period character to tell the model to stop once it finishes a sentence.
Being Creative vs. Predictable
Probably the most useful set of parameters are the ones that we can tune to control the randomness of the output. The beauty of working with LLMs is, for the same prompt, the next generated token will not be the same every time. Rather, it is sampled from a long list of possible tokens. This is where the creative aspect of LLM comes from, allowing us to generate a variety of outputs, given the exact same prompt.
But depending on your application, you may want to reduce, or increase, this level of randomness. You can do this by adjusting a number of parameters.
But before looking at the parameters, it’s worth taking the time to understand how the model selects the next token to generate. It does this by assigning a likelihood
number to each of all possible next tokens. The model would see that the token cookies
has a much higher likelihood than chair
for appearing after the phrase I like to bake.
During text generation, there’s still a probability that chair
would appear, but the probability is much lower than cookies.
The parameters we are going to see now can change this behavior.
There are three parameters we can adjust for this.
RANDOMNESS
— is how the model chooses from its next choice of tokens. Lower temperature will cause the model to output text that is more predictable, while higher temperature means that the output will be more creative. It is a number between 0 and 2, and in most cases, somewhere between 0 and 1 works fine.TOP-K
— is the list in which the model can make its choices from. The default is 0, which means the model will consider all possible next tokens out of the thousands of possible tokens. If you change to any other number, of say 100, the model will only consider the top most probable tokens, limited by the number you define.

Saving and Sharing Your Prompts
You can save the prompts that you have created by clicking on the Save
button on the top right of the Playground. Once you have saved a prompt, it will appear as a preset on the Playground.
You can also share these prompts with others. To do this, click on the Share
button on the top right of the Playground. You will get a link which you can share with anyone!
Conclusion
In this article, we covered how to prompt a model — probably the most important, and definitely the most fun part of working with large language models. If you’d like to dive deeper into it, here are some resources you can go to for further reading:
- Documentation on designing a prompt
- Documentation on text generation
- The Generate API reference
- More prompt tips and tricks
In Part 2, we will explore the range of use cases and areas where generative AI in language can be applied. And along the way, we’ll see the different ways a prompt can be designed.