From 8c22fc1437be9544f3a22ce93f834081152244c2 Mon Sep 17 00:00:00 2001 From: yanismiraoui Date: Fri, 12 Jun 2026 11:07:15 -0700 Subject: [PATCH 1/2] docs: add ChatInception chat model page Document the new Inception Labs chat model integration and register it in the chat models nav (SUMMARY.md) and node list (README.md). Co-Authored-By: Claude Opus 4.8 (1M context) --- en/SUMMARY.md | 1 + .../langchain/chat-models/README.md | 1 + .../langchain/chat-models/chatinception.md | 28 +++++++++++++++++++ 3 files changed, 30 insertions(+) create mode 100644 en/integrations/langchain/chat-models/chatinception.md diff --git a/en/SUMMARY.md b/en/SUMMARY.md index 1b006348..150cfce0 100644 --- a/en/SUMMARY.md +++ b/en/SUMMARY.md @@ -119,6 +119,7 @@ * [ChatGoogleGenerativeAI](integrations/langchain/chat-models/google-ai.md) * [Google VertexAI](integrations/langchain/chat-models/google-vertexai.md) * [ChatHuggingFace](integrations/langchain/chat-models/chathuggingface.md) + * [ChatInception](integrations/langchain/chat-models/chatinception.md) * [ChatLocalAI](integrations/langchain/chat-models/chatlocalai.md) * [ChatMistralAI](integrations/langchain/chat-models/mistral-ai.md) * [IBM Watsonx](integrations/langchain/chat-models/ibm-watsonx.md) diff --git a/en/integrations/langchain/chat-models/README.md b/en/integrations/langchain/chat-models/README.md index ba6c6378..c96d181a 100644 --- a/en/integrations/langchain/chat-models/README.md +++ b/en/integrations/langchain/chat-models/README.md @@ -20,6 +20,7 @@ Chat models take a list of messages as input and return a model-generated messag * [ChatGooglePaLM](broken-reference) * [Google VertexAI](google-vertexai.md) * [ChatHuggingFace](chathuggingface.md) +* [ChatInception](chatinception.md) * [ChatLocalAI](chatlocalai.md) * [ChatMistralAI](mistral-ai.md) * [ChatOllama](chatollama.md) diff --git a/en/integrations/langchain/chat-models/chatinception.md b/en/integrations/langchain/chat-models/chatinception.md new file mode 100644 index 00000000..42f5aacd --- /dev/null +++ b/en/integrations/langchain/chat-models/chatinception.md @@ -0,0 +1,28 @@ +--- +description: Wrapper around Inception Mercury diffusion LLMs (OpenAI-compatible API). +--- + +# ChatInception + +## Description +[Inception](https://inceptionlabs.ai/) provides the Mercury family of diffusion LLMs, including **Mercury 2** — a fast reasoning model that supports tool calling and structured outputs. The API is OpenAI-compatible and served from `https://api.inceptionlabs.ai/v1`. + +## Prerequisite +1. Create an account on the [Inception Platform](https://platform.inceptionlabs.ai/). +2. Get your API key from the [Inception Platform dashboard](https://platform.inceptionlabs.ai/). New accounts include free tokens to get started. + +## Step by Step Guide +1. **Chat Models** > Drag the **ChatInception** node. +2. Create a new credential with your Inception API key. +3. Select a **Model Name** (defaults to `mercury-2`). +4. (Optional) Click **Additional Parameters** to configure **Reasoning Effort** (`instant`, `low`, `medium`, `high`), **Max Tokens**, **Streaming**, **Timeout**, or override the **Base Path**. + +## Reasoning Effort +Mercury 2 is a reasoning model. Use the **Reasoning Effort** parameter to trade latency for depth: + +| Value | Behavior | +| --- | --- | +| `instant` | Lowest latency, no extended thinking | +| `low` | Minimal reasoning | +| `medium` | Balanced reasoning (default) | +| `high` | Maximum reasoning for complex tasks | From 399e775d75af0fec6466b4400361fc604f541915 Mon Sep 17 00:00:00 2001 From: yanismiraoui Date: Fri, 12 Jun 2026 11:12:21 -0700 Subject: [PATCH 2/2] address gemini comment --- en/integrations/langchain/chat-models/chatinception.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/en/integrations/langchain/chat-models/chatinception.md b/en/integrations/langchain/chat-models/chatinception.md index 42f5aacd..5d9c3310 100644 --- a/en/integrations/langchain/chat-models/chatinception.md +++ b/en/integrations/langchain/chat-models/chatinception.md @@ -7,7 +7,7 @@ description: Wrapper around Inception Mercury diffusion LLMs (OpenAI-compatible ## Description [Inception](https://inceptionlabs.ai/) provides the Mercury family of diffusion LLMs, including **Mercury 2** — a fast reasoning model that supports tool calling and structured outputs. The API is OpenAI-compatible and served from `https://api.inceptionlabs.ai/v1`. -## Prerequisite +## Prerequisites 1. Create an account on the [Inception Platform](https://platform.inceptionlabs.ai/). 2. Get your API key from the [Inception Platform dashboard](https://platform.inceptionlabs.ai/). New accounts include free tokens to get started.