AI Prosumer
EN

Model creator

DeepSeek

deepseek/

Explore models by DeepSeek. Compare available providers, model tags and inference pricing in one place.

DeepSeek develops general-purpose, reasoning and coding models. Its DeepSeek V3.2 and V4 model releases publish model weights under the MIT license.

Models

Prices per 1M tokens

deepseek/deepseek-v2.5

An upgraded version of DeekSeek-V2 that integrates the general and coding abilities of both DeepSeek-V2-Chat and DeepSeek-Coder-V2-Instruct.

7 model tags164K contextToken exchange
Model details

deepseek/deepseek-r1

DeepSeek-R1 is a family of open reasoning models with performance approaching that of leading models, such as O3 and Gemini 2.5 Pro.

35 model tags131K contextToken exchange
Model details

deepseek/deepseek-v3

A strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token.

5 model tags164K contextToken exchange
Model details

deepseek/deepseek-v4-pro

DeepSeek-V4-Pro is a model by DeepSeek that accepts text and produces text. This catalog entry identifies deepseek/deepseek-v4-pro. Its reviewed context window is 1,048,576 tokens; deployment limits can be lower.

* Credits availableAvailable from ShareAI1M contextFrom €1.568779/M inputFrom €3.137558/M output
Model details

deepseek/deepseek-v4-flash

DeepSeek-V4-Flash is a model by DeepSeek that accepts text and produces text. This catalog entry identifies deepseek/deepseek-v4-flash. Its reviewed context window is 1,048,576 tokens; deployment limits can be lower.

* Credits availableAvailable from ShareAI1M contextFrom €0.171303/M inputFrom €0.459815/M output
Model details