We use cookies. Cookies and browser storage keep you signed in and remember your preferences. With your permission we also measure how the site is used. You can any time. See our cookie policy.
ricursiveai/ripple-65b-previewRuns on your balance at the prices above. New accounts start with free usage.
Settings
Ripple-65B-Preview is a text model from ricursiveAI, released as a preview build. It is a 65B parameter model served in BF16, and it sits in the text category, so it is aimed at language work rather than image, audio or other modalities. The preview label indicates this is an early version of the model rather than a final, settled release.
The model takes text as input and returns text as output, with a context window of 1,048,576 tokens. That is enough room to hold very large inputs in a single request, such as long documents, extensive code files or lengthy conversation histories, without splitting them across calls. The maximum output is 131,072 tokens, so responses can be long as well. Native tool calling is part of the model itself, which means it can decide when to call functions you define and incorporate the results into its answer.
Those capabilities point to a few practical patterns. The large context suits tasks where the model needs to reason over a whole corpus at once: reviewing a long contract, summarising a large codebase, or answering questions against a big set of documents without a separate retrieval step. Native tool calling fits agent-style workflows, where the model queries an API, a database or an internal service and then continues from the returned data. The long output limit helps when you need a full report, a large refactor or a complete file rather than a short reply.
You can call it through api14's OpenAI-compatible API using the model id "ricursiveai/ripple-65b-preview". Because the interface follows the OpenAI format, existing OpenAI SDKs and clients work with a change of base URL and model id. The request parameters it supports are logprobs, max_tokens, response_format, seed, temperature, tool_choice, tools, top_logprobs and top_p. That covers sampling control through temperature and top_p, deterministic runs through seed, structured output through response_format, and function calling through tools and tool_choice. Logprobs and top_logprobs are available if you need token-level confidence information.
Since this is a preview release, it is worth testing against your own workloads before committing to it in production, and checking whether a newer build has appeared. If your task is small or latency-sensitive, a 65B model may be more than you need, and a smaller model could serve you better. If your work is not text-based, this model is not the right fit, since it only accepts and produces text.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api14.com/api/v1",
apiKey: process.env.API14_API_KEY,
});
const result = await client.chat.completions.create({
"model": "ricursiveai/ripple-65b-preview",
"messages": [
{
"role": "user",
"content": "Explain vector databases in two sentences."
}
]
});
console.log(result.choices[0].message.content);
logprobsmax_tokensresponse_formatseedtemperaturetool_choicetoolstop_logprobstop_pThis community has no topics yet. Open the first one about Ricursive: Ripple 65B Preview.