Skip to content

Miscellaneous Chat template improvements #1872

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 7 commits into from
Feb 28, 2025
Merged

Miscellaneous Chat template improvements #1872

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
5134b19
Encourage async in response generation
cpsievert Feb 26, 2025
c692cca
Update requirements and update models
cpsievert Feb 26, 2025
8f1d234
Add template.env files
cpsievert Feb 26, 2025
e5b55e1
Remove unnecessary call to markdown() in hello chat template
cpsievert Feb 27, 2025
22037aa
Proper naming of hello chat express app file
cpsievert Feb 27, 2025
d3f4c93
Add next steps and follow up
cpsievert Feb 27, 2025
d880543
Merge branch 'main' into chat-template-updates
cpsievert Feb 27, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 15 additions & 1 deletion shiny/templates/chat/llm-enterprise/aws-bedrock-anthropic/_template.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,19 @@
{
"type": "app",
"id": "chat-ai-anthropic-aws",
"title": "Chat AI using Anthropic via AWS Bedrock"
"title": "Chat AI using Anthropic via AWS Bedrock",
"next_steps": [
"Put your Bedrock credentials in the `template.env` file and rename it to `.env`.",
"Run the app with `shiny run app.py`."
],
"follow_up": [
{
"type": "info",
"text": "Need help connecting to Bedrock?"
},
{
"type": "action",
"text": "Learn more at https://posit-dev.github.io/chatlas/reference/ChatBedrockAnthropic.html"
}
]
}
6 changes: 1 addition & 5 deletions shiny/templates/chat/llm-enterprise/aws-bedrock-anthropic/app.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,10 +15,6 @@
load_dotenv()
chat_model = ChatBedrockAnthropic(
model="anthropic.claude-3-sonnet-20240229-v1:0",
# aws_secret_key=os.getenv("AWS_SECRET_KEY"),
# aws_access_key=os.getenv("AWS_ACCESS_KEY"),
# aws_region=os.getenv("AWS_REGION"),
# aws_account_id=os.getenv("AWS_ACCOUNT_ID"),
)

# Set some Shiny page options
Expand All @@ -36,5 +32,5 @@
# Define a callback to run when the user submits a message
@chat.on_user_submit
async def handle_user_input(user_input: str):
response = chat_model.stream(user_input)
response = await chat_model.stream_async(user_input)
await chat.append_message_stream(response)
1 change: 0 additions & 1 deletion shiny/templates/chat/llm-enterprise/aws-bedrock-anthropic/requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,4 @@
shiny
python-dotenv
tokenizers
chatlas
anthropic[bedrock]
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Just a note that reading through this PR reminded me of an issue I opened in chatlas to make provider-specific extras in chatlas. Here's the related PR: posit-dev/chatlas#66

In the future (I guess not yet...) it'd be nice to be able to include those extras in these templates -- e.g. chatlas[bedrock-anthropic] -- which would mean that updates to the provider dependencies don't require changes here to stay up-to-date.

Copy link
Collaborator Author

@cpsievert cpsievert Feb 28, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yea, agreed, thanks for the PR

FWIW, tokenizers was once for ui.Chat.messages(token_limits), but we longer advertise it

6 changes: 6 additions & 0 deletions shiny/templates/chat/llm-enterprise/aws-bedrock-anthropic/template.env
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
# Once you provided your API key, rename this file to .env
# The load_dotenv() in the app.py will then load this env variable
Comment on lines +1 to +2
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think it'd be good to include this message in the shiny create output. See the template.json in the shiny/templates/apps section for examples. Here's one:

{
  "type": "app",
  "id": "dashboard",
  "title": "Basic dashboard",
  "description": "A basic, single page dashboard with value boxes, two plots in cards and a sidebar.",
  "next_steps": [
    "Run the app with `shiny run app.py`."
  ],
  "follow_up": [
    {
      "type": "info",
      "text": "Just getting started with Shiny?"
    },
    {
      "type": "action",
      "text": "Learn more at https://shiny.posit.co/py/docs/user-interfaces.html"
    }
  ]
}

Copy link
Collaborator Author

@cpsievert cpsievert Feb 27, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good call, forgot that was a thing!

AWS_SECRET_KEY=<Your AWS info here>
AWS_ACCESS_KEY=<Your AWS info here>
AWS_REGION=<Your AWS info here>
AWS_ACCOUNT_ID=<Your AWS info here>
16 changes: 15 additions & 1 deletion shiny/templates/chat/llm-enterprise/azure-openai/_template.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,19 @@
{
"type": "app",
"id": "chat-ai-azure-openai",
"title": "Chat AI using OpenAI via Azure"
"title": "Chat AI using OpenAI via Azure",
"next_steps": [
"Put your Azure credentials in the `template.env` file and rename it to `.env`.",
"Run the app with `shiny run app.py`."
],
"follow_up": [
{
"type": "info",
"text": "Need help connecting to Azure?"
},
{
"type": "action",
"text": "Learn more at https://posit-dev.github.io/chatlas/reference/ChatAzureOpenAI.html"
}
]
}
2 changes: 1 addition & 1 deletion shiny/templates/chat/llm-enterprise/azure-openai/app.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,5 +37,5 @@
# Define a callback to run when the user submits a message
@chat.on_user_submit
async def handle_user_input(user_input: str):
response = chat_model.stream(user_input)
response = await chat_model.stream_async(user_input)
await chat.append_message_stream(response)
1 change: 0 additions & 1 deletion shiny/templates/chat/llm-enterprise/azure-openai/requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,4 @@
shiny
python-dotenv
tokenizers
chatlas
openai
3 changes: 3 additions & 0 deletions shiny/templates/chat/llm-enterprise/azure-openai/template.env
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
# Once you provided your API key, rename this file to .env
# The load_dotenv() in the app.py will then load this env variable
AZURE_OPENAI_API_KEY=<Your Azure OpenAI API key>
16 changes: 15 additions & 1 deletion shiny/templates/chat/llms/anthropic/_template.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,19 @@
{
"type": "app",
"id": "chat-ai-anthropic",
"title": "Chat AI using Anthropic"
"title": "Chat AI using Anthropic",
"next_steps": [
"Put your Anthropic API key in the `template.env` file and rename it to `.env`.",
"Run the app with `shiny run app.py`."
],
"follow_up": [
{
"type": "info",
"text": "Need help obtaining an API key?"
},
{
"type": "action",
"text": "Learn how to obtain one at https://posit-dev.github.io/chatlas/reference/ChatAnthropic.html"
}
]
}
4 changes: 2 additions & 2 deletions shiny/templates/chat/llms/anthropic/app.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@
load_dotenv()
chat_model = ChatAnthropic(
api_key=os.environ.get("ANTHROPIC_API_KEY"),
model="claude-3-5-sonnet-latest",
model="claude-3-7-sonnet-latest",
system_prompt="You are a helpful assistant.",
)

Expand All @@ -37,5 +37,5 @@
# Generate a response when the user submits a message
@chat.on_user_submit
async def handle_user_input(user_input: str):
response = chat_model.stream(user_input)
response = await chat_model.stream_async(user_input)
await chat.append_message_stream(response)
3 changes: 3 additions & 0 deletions shiny/templates/chat/llms/anthropic/template.env
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
# Once you provided your API key, rename this file to .env
# The load_dotenv() in the app.py will then load this env variable
ANTHROPIC_API_KEY=<Your Anthropic API key>
16 changes: 15 additions & 1 deletion shiny/templates/chat/llms/google/_template.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,19 @@
{
"type": "app",
"id": "chat-ai-gemini",
"title": "Chat AI using Google Gemini"
"title": "Chat AI using Google Gemini",
"next_steps": [
"Put your Google API key in the `template.env` file and rename it to `.env`.",
"Run the app with `shiny run app.py`."
],
"follow_up": [
{
"type": "info",
"text": "Need help obtaining an API key?"
},
{
"type": "action",
"text": "Learn how to obtain one at https://posit-dev.github.io/chatlas/reference/ChatGoogle.html"
}
]
}
4 changes: 2 additions & 2 deletions shiny/templates/chat/llms/google/app.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@
chat_model = ChatGoogle(
api_key=os.environ.get("GOOGLE_API_KEY"),
system_prompt="You are a helpful assistant.",
model="gemini-1.5-flash",
model="gemini-2.0-flash",
)

# Set some Shiny page options
Expand All @@ -33,5 +33,5 @@
# Generate a response when the user submits a message
@chat.on_user_submit
async def handle_user_input(user_input: str):
response = chat_model.stream(user_input)
response = await chat_model.stream_async(user_input)
await chat.append_message_stream(response)
5 changes: 2 additions & 3 deletions shiny/templates/chat/llms/google/requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,4 @@
shiny
python-dotenv
tokenizers
chatlas
google-generativeai
chatlas>=0.4.0
google-genai
3 changes: 3 additions & 0 deletions shiny/templates/chat/llms/google/template.env
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
# Once you provided your API key, rename this file to .env
# The load_dotenv() in the app.py will then load this env variable
GOOGLE_API_KEY=<Your Google API key>
6 changes: 5 additions & 1 deletion shiny/templates/chat/llms/langchain/_template.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,9 @@
{
"type": "app",
"id": "chat-ai-langchain",
"title": "Chat AI using LangChain"
"title": "Chat AI using LangChain",
"next_steps": [
"Put your OpenAI API key in the `template.env` file and rename it to `.env`.",
"Run the app with `shiny run app.py`."
]
}
2 changes: 1 addition & 1 deletion shiny/templates/chat/llms/langchain/app.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -38,5 +38,5 @@
# Define a callback to run when the user submits a message
@chat.on_user_submit
async def handle_user_input(user_input: str):
response = chat_model.stream(user_input)
response = await chat_model.stream_async(user_input)
await chat.append_message_stream(response)
1 change: 0 additions & 1 deletion shiny/templates/chat/llms/langchain/requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,3 @@
shiny
python-dotenv
tokenizers
langchain-openai
3 changes: 3 additions & 0 deletions shiny/templates/chat/llms/langchain/template.env
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
# Once you provided your API key, rename this file to .env
# The load_dotenv() in the app.py will then load this env variable
OPENAI_API_KEY=<Your OpenAI API key>
17 changes: 16 additions & 1 deletion shiny/templates/chat/llms/ollama/_template.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,20 @@
{
"type": "app",
"id": "chat-ai-ollama",
"title": "Chat AI using Ollama"
"title": "Chat AI using Ollama",
"next_steps": [
"If you haven't already, download the Ollama executable from https://ollama.com/",
"Run the executable and download the relevant model (llama3.2)",
"Run the app with `shiny run app.py`."
],
"follow_up": [
{
"type": "info",
"text": "Need help connecting to Ollama?"
},
{
"type": "action",
"text": "Learn more at https://posit-dev.github.io/chatlas/reference/ChatOllama.html"
}
]
}
4 changes: 2 additions & 2 deletions shiny/templates/chat/llms/ollama/app.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,7 @@
# ChatOllama() requires an Ollama model server to be running locally.
# See the docs for more information on how to set up a local Ollama server.
# https://posit-dev.github.io/chatlas/reference/ChatOllama.html
chat_model = ChatOllama(model="llama3.1")
chat_model = ChatOllama(model="llama3.2")

# Set some Shiny page options
ui.page_opts(
Expand All @@ -29,5 +29,5 @@
# Generate a response when the user submits a message
@chat.on_user_submit
async def handle_user_input(user_input: str):
response = chat_model.stream(user_input)
response = await chat_model.stream_async(user_input)
await chat.append_message_stream(response)
1 change: 0 additions & 1 deletion shiny/templates/chat/llms/ollama/requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,4 +1,3 @@
shiny
tokenizers
chatlas
ollama
16 changes: 15 additions & 1 deletion shiny/templates/chat/llms/openai/_template.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,19 @@
{
"type": "app",
"id": "chat-ai-openai",
"title": "Chat AI using OpenAI"
"title": "Chat AI using OpenAI",
"next_steps": [
"Put your OpenAI API key in the `template.env` file and rename it to `.env`.",
"Run the app with `shiny run app.py`."
],
"follow_up": [
{
"type": "info",
"text": "Need help obtaining an API key?"
},
{
"type": "action",
"text": "Learn how to obtain one at https://posit-dev.github.io/chatlas/reference/ChatOpenAI.html"
}
]
}
2 changes: 1 addition & 1 deletion shiny/templates/chat/llms/openai/app.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,5 +37,5 @@
# Generate a response when the user submits a message
@chat.on_user_submit
async def handle_user_input(user_input: str):
response = chat_model.stream(user_input)
response = await chat_model.stream_async(user_input)
await chat.append_message_stream(response)
1 change: 0 additions & 1 deletion shiny/templates/chat/llms/openai/requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,4 @@
shiny
python-dotenv
tokenizers
chatlas
openai
3 changes: 3 additions & 0 deletions shiny/templates/chat/llms/openai/template.env
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,3 @@
# Once you provided your API key, rename this file to .env
# The load_dotenv() in the app.py will then load this env variable
OPENAI_API_KEY=<Your OpenAI API key>
16 changes: 15 additions & 1 deletion shiny/templates/chat/llms/playground/_template.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,19 @@
{
"type": "app",
"id": "chat-ai-playground",
"title": "Chat Playground w/ OpenAI, Anthropic, and Google"
"title": "Chat Playground w/ OpenAI, Anthropic, and Google",
"next_steps": [
"Put your OpenAI, Anthropic, and Google API key in the `template.env` file and rename it to `.env`.",
"Run the app with `shiny run app.py`."
],
"follow_up": [
{
"type": "info",
"text": "Need help obtaining an API keys?"
},
{
"type": "action",
"text": "Learn how to obtain them at https://posit-dev.github.io/chatlas/reference/"
}
]
}
6 changes: 3 additions & 3 deletions shiny/templates/chat/llms/playground/app.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,13 +15,13 @@
load_dotenv()

models = {
"openai": ["gpt-4o-mini", "gpt-4o"],
"claude": [
"claude-3-7-sonnet-latest",
"claude-3-opus-latest",
"claude-3-5-sonnet-latest",
"claude-3-haiku-20240307",
],
"google": ["gemini-1.5-pro-latest"],
"openai": ["gpt-4o-mini", "gpt-4o"],
"google": ["gemini-2.0-flash"],
}

model_choices: dict[str, dict[str, str]] = {}
Expand Down
4 changes: 2 additions & 2 deletions shiny/templates/chat/llms/playground/requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
chatlas
chatlas>=0.4
openai
anthropic
google-generativeai
google-genai
python-dotenv
shiny
5 changes: 5 additions & 0 deletions shiny/templates/chat/llms/playground/template.env
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
# Once you provided your API key, rename this file to .env
# The load_dotenv() in the app.py will then load this env variable
ANTHROPIC_API_KEY=<Your Anthropic API key>
OPENAI_API_KEY=<Your OpenAI API key>
GOOGLE_API_KEY=<Your Google API key>
5 changes: 4 additions & 1 deletion shiny/templates/chat/starters/hello/_template.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,8 @@
{
"type": "app",
"id": "chat-hello",
"title": "Hello Shiny Chat"
"title": "Hello Shiny Chat",
"next_steps": [
"Run the app with `shiny run app.py`."
]
}
12 changes: 5 additions & 7 deletions shiny/templates/chat/starters/hello/app-core.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,13 +7,11 @@
)

# Create a welcome message
welcome = ui.markdown(
"""
Hi! This is a simple Shiny `Chat` UI. Enter a message below and I will
simply repeat it back to you. For more examples, see this
[folder of examples](https://github.com/posit-dev/py-shiny/tree/main/shiny/templates/chat).
"""
)
welcome = """
Hi! This is a simple Shiny `Chat` UI. Enter a message below and I will
simply repeat it back to you. For more examples, see this
[folder of examples](https://github.com/posit-dev/py-shiny/tree/main/shiny/templates/chat).
"""


def server(input, output, session):
Expand Down
Loading
Loading