Cohere: Rerank 4 Fast

Overview

Lower-latency version of Rerank 4 for high-throughput search.

Provider

Cohere

Modality

Rerank

Input

text

Output

text

Weights

Closed

Distillable

NoLicence or terms, checked October 8, 2026.

Tags

rerank, retrieval, low-cost

Docs

Provider documentation

Data policy

Prompts go to

Cohere SaaS Platform

Retention

Logged prompts and generations deleted after 30 days unless flagged or legally required, you can opt out of training use in dashboard Data Controls, ZDR for approved enterprise customers

Zero data retention

Available

Region

Not stated

Policy

Cohere data policy

Summarised from the provider's published terms. The provider's own policy is what applies.

Where you can run it

How Rerank 4 Fast can be deployed, bought and reached through Safeguard.

Deployment

Public cloudPrivate cloud

Procurement

Not published yet

Inference regions

🇧🇷Brazil🇩🇪Germany🇬🇧United Kingdom🇮🇳India🇯🇵Japan🇸🇦Saudi Arabia🇺🇸United StatesCloud provider's availability page, checked October 8, 2026.

Closed weights stay with the provider, so this model runs in the provider's cloud, never on-prem or air-gapped.

Pricing

Notes

Billed per search: $2.00 per 1,000 searches (one query with up to 100 documents). Not token-priced, so per-1M fields are null.

Source

Cohere pricing , last verified October 8, 2026

No markup on model usage: you pay the provider's list price.

API

OpenAI-compatible. Your API key and access to this model are set up for your workspace when you choose Use with Safeguard.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.safeguard.sh/gpt/v1",
    api_key="SAFEGUARD_API_KEY",
)

response = client.chat.completions.create(
    model="cohere/rerank-v4.0-fast",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Base URL https://api.safeguard.sh/gpt/v1, model cohere/rerank-v4.0-fast.