Source: vercel.com/changelog/muse-spark-open-now-available-on-ai-gateway
Meta's open-weight reasoning model, Muse Spark Open, is now available through Vercel's AI Gateway. Vercel positions it for coding and agentic work, including multi-file refactoring, debugging and reading large codebases. It accepts text and image input with a 1M-token context window, so code and screenshots can share a single request.
Developers select the model with the name meta/muse-spark-open across the AI SDK, the OpenAI-compatible Chat Completions and Responses APIs, the Anthropic-compatible Messages API, and coding agents connected to the gateway. Vercel says the model is available with zero data retention and no training, without a separate provider account. For coding agents, users upgrade the Vercel CLI, run the coding agents setup command, and then choose the model inside the agent.
AI Gateway also provides a unified API for calling models, tracking usage and cost, and configuring retries and failover. It includes routing rules and budgets for API keys. Vercel says it passes provider pricing through with no markup and charges no platform fee on inference, including bring-your-own-key requests.
Why it matters if you build agents
Agent builders can reach this model through an existing AI SDK, OpenAI-compatible or Anthropic-compatible setup by changing the model name, without managing a separate Meta account. Teams with data handling requirements should compare the zero data retention and no-training terms against their own policies. Because the model accepts images, screenshots and code can go into the same request during debugging tasks.
Get stories like this every weekday
One short email with the day's agent tooling news.