Add TokenMix chat model node

[TokenMixAi]

Description

Adds a TokenMix chat model node. TokenMix is an OpenAI-compatible API gateway that exposes DeepSeek, Qwen, Kimi, GLM, MiniMax and other models through a single endpoint (https://api.tokenmix.ai/v1) and one API key.

Because the gateway is OpenAI-compatible, the node wraps @langchain/openai's ChatOpenAI with the TokenMix base URL — the same approach as the existing OpenRouter node. It includes vision (image upload) support via IVisionChatModal, configurable base path, streaming, and the standard sampling parameters.

New files

• packages/components/nodes/chatmodels/ChatTokenMix/ChatTokenMix.ts — node definition
• packages/components/nodes/chatmodels/ChatTokenMix/FlowiseChatTokenMix.ts — ChatOpenAI subclass with multimodal support
• packages/components/nodes/chatmodels/ChatTokenMix/tokenmix.svg — node icon
• packages/components/credentials/TokenMixApi.credential.ts — API key credential

Nodes and credentials are auto-discovered, so no registry changes are required.

How it works

The user supplies a TokenMix API key (credential tokenMixApi) and a model name (e.g. deepseek/deepseek-v4-pro). The node points ChatOpenAI's configuration.baseURL at https://api.tokenmix.ai/v1.

Local checks

$ prettier --config package.json --check \
    packages/components/credentials/TokenMixApi.credential.ts \
    packages/components/nodes/chatmodels/ChatTokenMix/*.ts
All matched files use Prettier code style!

The node mirrors the existing ChatOpenRouter node structure (same imports and type surface), with no new dependencies.

[gemini-code-assist]

Code Review

This pull request introduces a new ChatTokenMix chat model node and its corresponding TokenMix API credential configuration, enabling integration with the TokenMix OpenAI-compatible Inference API. The review feedback points out a potential runtime issue where an empty or missing temperature parameter could evaluate to NaN and cause API errors. It is recommended to conditionally assign the temperature property only when it is a valid, non-empty value.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[gemini-code-assist]

If the optional temperature parameter is not provided or is left empty, parseFloat(temperature) will evaluate to NaN. Passing NaN as the temperature to the ChatOpenAI constructor will cause API errors or unexpected behavior at runtime.

To prevent this, only assign the temperature property if it is a valid number/string and not nullish or empty, adhering to the loose equality standard for nullish checks.

Suggested change

	const obj: ChatOpenAIFields = {
	temperature: parseFloat(temperature),
	modelName,
	openAIApiKey: tokenMixApiKey,
	apiKey: tokenMixApiKey,
	streaming: streaming ?? true
	}
	const obj: ChatOpenAIFields = {
	modelName,
	openAIApiKey: tokenMixApiKey,
	apiKey: tokenMixApiKey,
	streaming: streaming ?? true
	}
	if (temperature != null && temperature !== '') {
	obj.temperature = parseFloat(temperature)
	}

References

1. In JavaScript/TypeScript, use loose equality (== null or != null) as a standard idiom for a 'nullish' check that covers both null and undefined.