---
title: "Models"
output: rmarkdown::html_vignette
vignette: >
%\VignetteIndexEntry{Models}
%\VignetteEngine{knitr::rmarkdown}
%\VignetteEncoding{UTF-8}
---
```{r, include = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>",
eval = FALSE
)
```
Under the hood, `querychat` is powered by [ellmer](https://ellmer.tidyverse.org/), a library for building chat-based applications with large language models (LLMs). `ellmer` supports a wide range of LLM providers -- [see here](https://ellmer.tidyverse.org/reference/index.html#chat-constructors) for a full list.
```{r}
library(querychat)
library(palmerpenguins)
library(ellmer)
```
## Specify a model
To use a particular model, pass a `"{provider}/{model}"` string to the `client` parameter, which gets passed along to `ellmer::chat()`:
```{r}
qc <- querychat(penguins, client = "anthropic/claude-sonnet-4-5")
qc$app() # Launch the app
```
And, if you'd like to effectively set a new default model, you can use the `querychat.client` R option or the `QUERYCHAT_CLIENT` environment variable.
```{r}
# In your .Rprofile
options(querychat.client = "anthropic/claude-sonnet-4-5")
```
Note that it can also be useful to pass a full `Chat` object to the `client` parameter for more advanced use cases (e.g., custom parameters, tools, etc):
```{r}
client <- chat_anthropic(model = "claude-sonnet-4-5")
qc <- querychat(penguins, client = client)
qc$app() # Launch the app
```
## Credentials
Most models require an API key or some other form of authentication. See the reference page for the relevant model provider (e.g., [`chat_anthropic()`](https://ellmer.tidyverse.org/reference/chat_anthropic.html)) to learn more on how to set up credentials.
::: {.alert .alert-info}
**GitHub model marketplace**
If you are already setup with GitHub credentials, [GitHub model marketplace](https://github.com/marketplace/models) provides a free and easy way to get started. See [here](https://ellmer.tidyverse.org/reference/chat_github.html) for more details on how to get setup.
```{r}
library(ellmer)
# Just works if GITHUB_TOKEN is set in your environment
client <- chat_github(model = "gpt-4.1")
```
:::
In general, most providers will prefer credentials stored as environment variables. Common practice is to use an `.Renviron` file to manage these variables. For example, for `chat_openai()`, you might add to your `.Renviron` file:
```bash
OPENAI_API_KEY="your_api_key_here"
```
Then, you can edit your `.Renviron` file using:
```{r}
usethis::edit_r_environ()
```
## Recommended models
In theory, you could use any model that has tool calling support, but we currently recommend (as of November 2025):
- GPT-4.1 (the default)
- Claude 4.5 Sonnet
- Google Gemini 3.0
In our testing, we've found that those models strike a good balance between accuracy and latency. That said, smaller/faster models like GPT-4.1-mini or Claude Haiku 4.5 work well for most tables and are worth trying first—they're significantly cheaper and faster. You can always switch to a larger model if you find the results aren't meeting your needs. On the other end of the spectrum, reasoning models like o3-mini tend to slow down responses without providing meaningfully better results for this task.
We've also seen some decent results with frontier local models (e.g., `gpt-oss:20b`), but even if you have the compute to run the largest models, they still tend to lag behind the cloud-hosted options in terms of accuracy and speed.
::: {.alert .alert-info}
**Data privacy concerns?**
If you have data privacy concerns, consider that your org may provide access to private instances of these models with data residency guarantees. For example, Azure, AWS Bedrock, and Google Vertex AI all provide private instances of popular LLMs. You can interface with these enterprise providers by passing the right string (e.g., `"bedrock-anthropic"`) or `Chat` object (e.g., `chat_bedrock_anthropic()`) to the `client` parameter. See the [ellmer docs](https://ellmer.tidyverse.org/reference/index.html#chat-constructors) for more details.
:::