IPWorks LLM beta combines model integration with content credentials

/n software has made IPWorks LLM 2026 Beta available, offering application developers components for the OpenAI, Anthropic, and Google Gemini APIs alongside tools for signing and verifying C2PA Content Credentials. The library also supports services implementing those APIs, including Ollama, vLLM, DeepSeek, Groq, and Azure OpenAI.

A provider-neutral LLMClient maintains conversations, tools, and replies in a common format, allowing applications to change providers through a single property or continue an existing conversation with a different provider. Supported capabilities include streaming responses, tool calling, text, image and file inputs, and structured output constrained to a JSON Schema. Embeddings, input token counting, and model listing are available where supported by the provider. Separate native components expose each provider’s API directly and can connect to compatible gateways, proxies, or locally hosted models.

The C2PASigner component embeds signed manifests into JPEG, PNG, and PDF files, recording who produced an asset, which tool was used, and whether generative AI was involved. C2PAVerifier checks signatures, certificate trust, content integrity, and revocation status. The toolkit is offered across platforms and languages including .NET, Java, C++, JavaScript, Python, Go, and Rust, although component availability varies by edition.