Text Generation ModelDeprecated

Sonar Large Chat

Sonar Large Chat is a text generation model from Perplexity built on Llama 3.1 with a 127,072-token context window.

Start Building with Sonar Large Chat View All Models

Publisher

Perplexity

TypeText

Context Window127,072 tokens

Replaced bySonar

Try Sonar Large Chat →

About Sonar Large Chat

Large-context chat with optional citations

Sonar Large Chat, formally named llama-3.1-sonar-large-128k-chat, is a chat-optimized text generation model published by Perplexity AI. It is built on Meta's Llama 3.1 architecture and supports a context window of up to 127,072 tokens with a maximum response size of 32,768 tokens. The model is part of Perplexity's Sonar family, which was designed to improve on earlier Sonar versions in cost-efficiency, speed, and overall performance.

Sonar Large Chat is suited for conversational tasks that benefit from long-context handling, such as document-grounded Q&A, summarization of lengthy inputs, and extended multi-turn dialogue. A notable feature is its configurable support for returning citations and images alongside responses, making it useful in workflows where source attribution matters. Note that this model has been marked as deprecated on MindStudio, so developers starting new projects may want to evaluate whether a successor model better fits their needs.

Capabilities

What Sonar Large Chat supports

Long Context Window

Processes up to 127,072 tokens in a single request, enabling analysis of lengthy documents or extended conversation histories.

Citation Return

Optionally returns source citations alongside generated responses, configurable via the Return Citations input.

Image Return

Can optionally return images alongside text responses when enabled via the Return Images input selector.

Extended Response Output

Supports response outputs of up to 32,768 tokens, allowing for detailed and thorough answers in a single generation.

Chat Optimization

Designed specifically for multi-turn conversational use cases, following the chat-tuned variant of the Llama 3.1 Sonar family.

Ready to build with Sonar Large Chat?

Get Started Free

FAQ

Common questions about Sonar Large Chat

What is the context window size for Sonar Large Chat?

Sonar Large Chat supports a context window of 127,072 tokens, with a maximum response size of 32,768 tokens.

Is Sonar Large Chat still available to use?

The model is currently marked as deprecated on MindStudio. It was added to the catalog on May 9, 2024. Developers starting new projects should check whether an updated Sonar model is available.

What underlying model is Sonar Large Chat based on?

Sonar Large Chat is based on Meta's Llama 3.1 architecture. Its full model name is llama-3.1-sonar-large-128k-chat, published by Perplexity AI.

Does Sonar Large Chat support citations in its responses?

Yes. The model includes a configurable Return Citations input that, when enabled, appends source citations to generated responses.

What is the pricing for Sonar Large Chat?

Pricing information is not published in the available metadata for this model. Check Perplexity's official documentation or MindStudio's pricing page for current rate details.

Resources