ChatTogether
This page will help you get started with Together AI chat models. For detailed documentation of all ChatTogether features and configurations head to the API reference.
Together AI offers an API to query 50+ leading open-source models
Overview
Integration details
Class | Package | Local | Serializable | JS support | Package downloads | Package latest |
---|---|---|---|---|---|---|
ChatTogether | langchain-together | ❌ | beta | ✅ |
Model features
Tool calling | Structured output | JSON mode | Image input | Audio input | Video input | Token-level streaming | Native async | Token usage | Logprobs |
---|---|---|---|---|---|---|---|---|---|
✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ✅ | ✅ |
Setup
To access Together models you'll need to create a/an Together account, get an API key, and install the langchain-together
integration package.
Credentials
Head to this page to sign up to Together and generate an API key. Once you've done this set the TOGETHER_API_KEY environment variable:
import getpass
import os
if "TOGETHER_API_KEY" not in os.environ:
os.environ["TOGETHER_API_KEY"] = getpass.getpass("Enter your Together API key: ")
If you want to get automated tracing of your model calls you can also set your LangSmith API key by uncommenting below:
# os.environ["LANGSMITH_API_KEY"] = getpass.getpass("Enter your LangSmith API key: ")
# os.environ["LANGSMITH_TRACING"] = "true"
Installation
The LangChain Together integration lives in the langchain-together
package:
%pip install -qU langchain-together
Instantiation
Now we can instantiate our model object and generate chat completions:
from langchain_together import ChatTogether
llm = ChatTogether(
model="meta-llama/Llama-3-70b-chat-hf",
temperature=0,
max_tokens=None,
timeout=None,
max_retries=2,
# other params...
)
Invocation
messages = [
(
"system",
"You are a helpful assistant that translates English to French. Translate the user sentence.",
),
("human", "I love programming."),
]
ai_msg = llm.invoke(messages)
ai_msg
AIMessage(content="J'adore la programmation.", response_metadata={'token_usage': {'completion_tokens': 9, 'prompt_tokens': 35, 'total_tokens': 44}, 'model_name': 'meta-llama/Llama-3-70b-chat-hf', 'system_fingerprint': None, 'finish_reason': 'stop', 'logprobs': None}, id='run-eabcbe33-cdd8-45b8-ab0b-f90b6e7dfad8-0', usage_metadata={'input_tokens': 35, 'output_tokens': 9, 'total_tokens': 44})
print(ai_msg.content)
J'adore la programmation.
Chaining
We can chain our model with a prompt template like so:
from langchain_core.prompts import ChatPromptTemplate
prompt = ChatPromptTemplate.from_messages(
[
(
"system",
"You are a helpful assistant that translates {input_language} to {output_language}.",
),
("human", "{input}"),
]
)
chain = prompt | llm
chain.invoke(
{
"input_language": "English",
"output_language": "German",
"input": "I love programming.",
}
)
AIMessage(content='Ich liebe das Programmieren.', response_metadata={'token_usage': {'completion_tokens': 7, 'prompt_tokens': 30, 'total_tokens': 37}, 'model_name': 'meta-llama/Llama-3-70b-chat-hf', 'system_fingerprint': None, 'finish_reason': 'stop', 'logprobs': None}, id='run-a249aa24-ee31-46ba-9bf9-f4eb135b0a95-0', usage_metadata={'input_tokens': 30, 'output_tokens': 7, 'total_tokens': 37})
API reference
For detailed documentation of all ChatTogether features and configurations head to the API reference: https://python.langchain.com/api_reference/together/chat_models/langchain_together.chat_models.ChatTogether.html