OpenClaw supports DeepSeek V4.1 Flash for text and image tasks
DeepSeek V4.1 Flash is available in OpenClaw, with image input, a million-token context window and support for agents using tools.
DeepSeek
OpenClaw supports DeepSeek V4.1 Flash as a model for its AI agents. Its documentation lists support for text and images, with capacity for one million tokens of context.
DeepSeek introduced V4.1 Flash on September 10, 2026, with image understanding built into the model. The company said its architecture reduces the memory and storage needed for cached context, which can affect the cost of running agents.
The OpenClaw integration supports reasoning across successive tool interactions, preserving information the model needs to continue a task. Users can connect a DeepSeek API key and select deepseek/deepseek-flash to use the model. OpenClaw also retains compatible older Flash selections, while recommending the current model selection for image inputs.
The day's biggest AI news in one short email, every morning. Plus an instant heads-up when a reset or a big release lands.