Try our new research platform with insights from 80,000+ expert users

Cohere vs DeepSeek comparison

 

Comparison Buyer's Guide

Executive Summary

Review summaries and opinions

We asked business professionals to review the solutions they use. Here are some excerpts of what they said:
 

Categories and Ranking

Cohere
Ranking in Large Language Models (LLMs)
5th
Average Rating
7.6
Reviews Sentiment
6.7
Number of Reviews
8
Ranking in other categories
AI Development Platforms (12th), AI Writing Tools (3rd), AI Proofreading Tools (5th)
DeepSeek
Ranking in Large Language Models (LLMs)
4th
Average Rating
3.0
Reviews Sentiment
3.7
Number of Reviews
1
Ranking in other categories
AI-Powered Chatbots (3rd)
 

Featured Reviews

AS
Engineer at Roche
Have improved project workflows using faster response times and reduced data embedding costs
One thing that Cohere can improve is related to some distances when I am trying similarity search. Let's suppose I have provided textual data that has been embedded. I have to use some extra process from numpy after embedding the model. In the case of OpenAI embedding models, I do not have to use that extra process, and they provide lower distances compared to my results from Cohere. I was getting distances of approximately 0.005 sometimes, but in the case of Cohere, I was getting distances around 0.5 or sometimes more than that. I think that can be improved. It was possibly because of some configuration or the way I was using it, but I am not exactly sure about that.
Malte Landwehr - PeerSpot reviewer
Vice President, Seo at idealo Internet GmbH
Experience frequent delays and privacy concerns with current research tool
I advise not to use the web interface or the mobile app. If you think you should use it, use the self-hosted version. I gave DeepSeek a rating of 3 out of 10 because the core product, when it works, is actually good. But so often it's not working that I just can't give it a higher score.

Quotes from Members

We asked business professionals to review the solutions they use. Here are some excerpts of what they said:
 

Pros

"Cohere helped us with all three aspects: money is saved, time is saved, and we needed fewer resources to meet our end goals."
"Cohere's Embed English v3.0 is a cloud-hosted model that took less time to embed the textual data and was more than 50 to 60% faster than other models, even somewhat faster than text-embedding-3 from OpenAI, helping to reduce development and embedding times."
"A key advantage of integrating Cohere’s reranking model is that it aligns with client requests to include a reranking module — a widely recognized method for improving RAG quality. Additionally, the API demonstrates strong performance in terms of response speed."
"The very first thing that I really like about it is the support team. They're really available on Discord, and they answer all of your questions."
"The best feature Cohere offers is the Reranking model."
"Cohere has positively impacted my organization by helping our customers work more efficiently when creating requests, and the embedding results are of very high quality."
"Cohere positively impacted my organization by improving the performance of my RAG system."
"Speed has helped me in my day-to-day work, and I really notice the difference because it responds very quickly to LLM requests."
"DeepSeek makes its reasoning traces public."
 

Cons

"Cohere could improve in areas where the command model is not as creative as some larger LLMs available in the market, which is expected but noticeable in open-ended generative tasks."
"One thing that Cohere can improve is related to some distances when I am trying similarity search."
"I believe Cohere can be improved technically by providing more feedback, logs, and metrics for embedding requests, as it currently appears to be a black box without any understanding of quality."
"It's challenging for us to make a conclusion about quality enhancement by using reranking models, as solid evaluation methodology for reranking is still immature."
"Cohere has text generation. I think it is mainly focused on AI search. If there was a way to combine the searches with images, I think it would be nice to include that."
"When performing similarity matching between text descriptions and the catalog descriptions created using Cohere, the matching could be improved."
"The documentation and support could be improved, as there is limited documentation available on the web."
"The product is often extremely slow, and queries frequently fail due to system overload."
report
Use our free recommendation engine to learn which Large Language Models (LLMs) solutions are best for your needs.
881,082 professionals have used our research since 2012.
 

Top Industries

By visitors reading reviews
Manufacturing Company
11%
Educational Organization
8%
Financial Services Firm
8%
University
7%
Educational Organization
17%
University
12%
Performing Arts
7%
Comms Service Provider
6%
 

Company Size

By reviewers
Large Enterprise
Midsize Enterprise
Small Business
By reviewers
Company SizeCount
Small Business2
Midsize Enterprise1
Large Enterprise6
No data available
 

Questions from the Community

What is your experience regarding pricing and costs for Cohere?
Compared to models available in the market, Cohere's pricing, setup cost, and licensing are better.
What needs improvement with Cohere?
Cohere could improve in areas where the command model is not as creative as some larger LLMs available in the market, which is expected but noticeable in open-ended generative tasks. Reporting and ...
What is your primary use case for Cohere?
We adopted Cohere primarily for their command model to support enterprise-grade text generation and NLP workflows. There was a use case for one of our customers where they required automated text g...
What needs improvement with DeepSeek?
I have data privacy concerns. The product is often extremely slow, and queries frequently fail due to system overload. DeepSeek needs to improve in stability, uptime, speed, and ensuring the tool i...
What is your primary use case for DeepSeek?
I use DeepSeek ( /products/deepseek-reviews ) for conducting research on complex questions and performing reasoning on complex questions. There have been countless instances where I needed to find ...
What advice do you have for others considering DeepSeek?
I advise not to use the web interface or the mobile app. If you think you should use it, use the self-hosted version. I gave DeepSeek a rating of 3 out of 10 because the core product, when it works...
 

Comparisons

 

Overview

Find out what your peers are saying about Google, OpenAI, Blackbox and others in Large Language Models (LLMs). Updated: December 2025.
881,082 professionals have used our research since 2012.