Amazon: Nova Pro 1.0

Amazon Nova Pro 1.0 is a text generation model from Amazon, listed on api14 under the model id amazon/nova-pro-v1. It sits in the text category and is described by Amazon as a capable multimodal model built to combine accuracy, speed, and cost across a wide range of tasks. It was released by the provider on 5 December 2024, and its knowledge cutoff is 31 October 2024, so anything that happened after that date is outside what the model was trained on.

On the input side, Nova Pro 1.0 accepts text and images, which makes it multimodal rather than text only. Output is text. The context window is 300,000 tokens, which is large enough to hold long documents, lengthy conversation histories, or a mix of text and images in a single request. The maximum output is 5,120 tokens, so responses are bounded well below the input capacity. If your task depends on producing very long single responses, that output ceiling is the number to keep in mind.

Those capabilities point to a fairly broad set of practical uses. Because it reads images as well as text, it fits tasks such as describing or reasoning about screenshots, diagrams, photos, and scanned pages alongside written instructions. The 300,000 token context suits long document analysis, summarisation of large files, and multi turn conversations where earlier context needs to stay in view. Tool support means it can be used in workflows where the model calls functions or external services, which is useful for assistants and agents. General text generation, extraction, and question answering over mixed text and image input are all reasonable fits.

You call the model through api14's OpenAI-compatible API using the model id "amazon/nova-pro-v1". The supported request parameters are max_tokens, stop, temperature, tools, top_k, and top_p. That covers the usual sampling controls (temperature, top_p, top_k), output length via max_tokens, stop sequences, and tool or function calling through tools. Because the interface is OpenAI-compatible, existing clients and SDKs that target the OpenAI chat format generally work with only the model id and base URL changed.

Two limits are worth weighing before you commit. The maximum output of 5,120 tokens is modest relative to the 300,000 token input window, so workloads that need very long generated answers may be constrained. The knowledge cutoff of 31 October 2024 also means the model has no built in awareness of later events, so tasks that depend on recent information need retrieval or tools rather than the model's own knowledge. If your work is text only and does not benefit from image input, a text only model may be a simpler fit, but for mixed text and image tasks with long context, Nova Pro 1.0 covers the ground described here.