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.
Cohere: Rerank v3.5
cohere/rerank-v3.5Try Cohere: Rerank v3.5 now
Runs on your balance at the prices above. New accounts start with free usage.
Results will appear here.
Settings
About Cohere: Rerank v3.5
Cohere: Rerank v3.5 is a reranking model from Cohere, listed under the rerank category. It is made by Cohere and is identified on api14 by the model id cohere/rerank-v3.5. Rather than generating free form text, the model is built to reorder search results so that the most relevant items appear first.
The model takes text as its input and produces a rerank output. Its context length is 4096 tokens, and the maximum output is 3686. Cohere describes Rerank v3.5 as designed to reorder search results for improved relevance, with support for multi-aspect and semi-structured data reranking across more than 100 languages. It works on text input and returns reranked results rather than prose, which shapes how you should expect to integrate it.
The practical use cases follow directly from that behaviour. Rerank v3.5 is suited to refining results from semantic or keyword search, where an initial retrieval step returns a candidate set and the reranker reorders those candidates by relevance. Because it handles multi-aspect and semi-structured data and covers over 100 languages, it can be used in multilingual search pipelines and in setups where records carry several fields or attributes that matter to relevance. Common patterns include improving the ordering of documents returned by a vector search or a keyword search before those results are shown to a user or passed to another step.
You call the model through api14's OpenAI-compatible API using the model id cohere/rerank-v3.5. According to the model metadata, the supported parameters list is empty, so no request parameters beyond what the API itself requires are documented for this model. Plan your integration around the model id and the text input to rerank output flow, and do not assume tunable sampling or similar parameters are available here.
Given the 4096 token context length and the rerank only output, this model fits retrieval and ordering tasks rather than generation, summarisation or chat. If you need a model that writes text, holds a long conversation or handles non text inputs, a different model on api14 would fit better.
How to call the Cohere: Rerank v3.5 API
- Create an API key in your dashboard.
- Install the official openai SDK for your language.
- Set baseURL to https://api14.com/api/v1 and model to "cohere/rerank-v3.5".
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api14.com/api/v1",
apiKey: process.env.API14_API_KEY,
});
const result = await client.post("/rerank", { body: {
"model": "cohere/rerank-v3.5",
"query": "What is the capital of France?",
"documents": [
"Paris is the capital of France.",
"Berlin is in Germany."
],
"top_n": 1
} });
console.log(result.results);
Specifications
- Context length
- 4,096
- Max output
- 3,686
- Input
- text
- Output
- rerank
Requests on api14 · last 7 days
Cohere: Rerank v3.5: frequently asked questions
- How much does the Cohere: Rerank v3.5 API cost?
- Cohere: Rerank v3.5 is free to call ($0 per token). Free variants have stricter upstream rate limits, so use them for testing and low-volume work.
- What is the context window of Cohere: Rerank v3.5?
- Cohere: Rerank v3.5 reads up to 4,096 tokens per request: prompt, history and documents together. It writes up to 3,686 output tokens per response.
- What can Cohere: Rerank v3.5 do?
- Cohere: Rerank v3.5 accepts text and returns rerank.
- How do I call Cohere: Rerank v3.5 from my code?
- Point any OpenAI SDK at https://api14.com/api/v1 and set the model to "cohere/rerank-v3.5". No other code changes. Copy a ready example from the code panel on this page.
- What are alternatives to Cohere: Rerank v3.5?
- Comparable Rerank models at a similar price: NVIDIA: Nemotron 3.5 Rerank 8B (free), VoyageAI by MongoDB: rerank-3-lite, VoyageAI by MongoDB: rerank-3. Each runs through the same API key, so switching is one string.
Cohere: Rerank v3.5 community
New topicThis community has no topics yet. Open the first one about Cohere: Rerank v3.5.