Workers AI - DeepSeek V4.1 Flash 1017 now available on Workers AI

<p><a href="https://developers.cloudflare.com/workers-ai/models/deepseek-v4.1-flash-1017/"><code>@cf/deepseek-ai/deepseek-v4.1-flash-1017</code></a> is now available on Workers AI. DeepSeek V4.1 Flash is a multimodal Mixture-of-Experts model with 552B backbone parameters. It activates 8B parameters per token during prefill and 16B during decode, which keeps input-heavy agentic workloads fast and low-cost.</p> <p>The 1017 release is an update to DeepSeek V4.1 Flash with improvements across reasoning, coding, agentic tasks, and visual benchmarks, and it completes tasks with shorter outputs.</p> <p><strong>Key capabilities:</strong></p> <ul> <li><strong>Vision</strong>: Send images alongside text in the same request.</li> <li><strong>Reasoning</strong>: Reasoning is on by default at <code>high</code> effort. Set <code>reasoning_effort</code> to <code>low</code> or <code>max</code> to change the effort, or to <code>none</code> to turn reasoning off.</li> <li><strong>Function calling</strong>: Build agents that invoke tools and APIs across multiple conversation turns.</li> <li><strong>Long context</strong>: A full 1,048,576 token context window.</li> </ul> <p>DeepSeek V4.1 Flash requires the <a href="https://developers.cloudflare.com/workers/platform/pricing/#workers">Workers Paid plan</a> or prepaid <a href="https://developers.cloudflare.com/ai-gateway/features/unified-billing/">AI Gateway credits</a>. The default rate limit is 20 requests per minute with standard Workers AI billing and 50 requests per minute with prepaid AI Gateway credits. For details, refer to <a href="https://developers.cloudflare.com/workers-ai/platform/limits/#paid-models">limits</a>.</p> <p>Use DeepSeek V4.1 Flash through the <a href="https://developers.cloudflare.com/workers-ai/configuration/bindings/">Workers AI binding</a> (<code>env.AI.run()</code>), the REST API, the <a href="https://developers.cloudflare.com/workers-ai/configuration/open-ai-compatibility/">OpenAI-compatible endpoint</a>, or <a href="https://developers.cloudflare.com/ai-gateway/">AI Gateway</a>.</p> <p>For more information, refer to the <a href="https://developers.cloudflare.com/workers-ai/models/deepseek-v4.1-flash-1017/">DeepSeek V4.1 Flash model page</a> and <a href="https://developers.cloudflare.com/workers-ai/platform/pricing/">pricing</a>.</p>