AI Agent Hub

LLM Models · deepseek-ai

Browse the world's large language models. Compare parameters, benchmarks, VRAM and more.

62 models from deepseek-ai Compare
DeepSeek-V3.1-Terminus logo
DeepSeek-V3.1-Terminus
deepseek-ai

This update maintains the model's original capabilities while addressing issues reported by users, including:

open-source ↓ 17K
deepseek-moe-16b-base logo
deepseek-moe-16b-base
deepseek-ai

[🏠Homepage] [🤖 Chat with DeepSeek LLM] [Discord] [Wechat(微信)]

open-source 16.4B ↓ 15.7K
deepseek-coder-1.3b-base logo
deepseek-coder-1.3b-base
deepseek-ai

[🏠Homepage] [🤖 Chat with DeepSeek Coder] [Discord] [Wechat(微信)]

code 1.3B ↓ 14.7K
DeepSeek-Coder-V2-Instruct logo
DeepSeek-Coder-V2-Instruct
deepseek-ai

DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence

code ↓ 9.9K
DeepSeek-Coder-V2-Lite-Base logo
DeepSeek-Coder-V2-Lite-Base
deepseek-ai

DeepSeek-Coder-V2: Breaking the Barrier of Closed-Source Models in Code Intelligence

code ↓ 9.5K
deepseek-vl-7b-chat logo
deepseek-vl-7b-chat
deepseek-ai

Introducing DeepSeek-VL, an open-source Vision-Language (VL) Model designed for real-world vision and language understanding applications. DeepSeek-VL possesses general multimodal understanding capabilities, capable of processing logical diagrams, web pages, formula recognition,…

multimodal 7.0B ↓ 9.4K
deepseek-llm-67b-base logo
deepseek-llm-67b-base
deepseek-ai

[🏠Homepage] [🤖 Chat with DeepSeek LLM] [Discord] [Wechat(微信)]

open-source 67.0B ↓ 9.1K
DeepSeek-R1-Zero logo
DeepSeek-R1-Zero
deepseek-ai

We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1. DeepSeek-R1-Zero, a model trained via large-scale reinforcement learning (RL) without supervised fine-tuning (SFT) as a preliminary step, demonstrated remarkable performance on reasoning. With R…

reasoning ↓ 8.2K
deepseek-vl-1.3b-chat logo
deepseek-vl-1.3b-chat
deepseek-ai

Introducing DeepSeek-VL, an open-source Vision-Language (VL) Model designed for real-world vision and language understanding applications. DeepSeek-VL possesses general multimodal understanding capabilities, capable of processing logical diagrams, web pages, formula recognition,…

multimodal 1.3B ↓ 8.1K
deepseek-vl2-small logo
deepseek-vl2-small
deepseek-ai

Introducing DeepSeek-VL2, an advanced series of large Mixture-of-Experts (MoE) Vision-Language Models that significantly improves upon its predecessor, DeepSeek-VL. DeepSeek-VL2 demonstrates superior capabilities across various tasks, including but not limited to visual question…

multimodal 16.0B ↓ 8K
DeepSeek-V2.5 logo
DeepSeek-V2.5
deepseek-ai

DeepSeek-V2.5 is an upgraded version that combines DeepSeek-V2-Chat and DeepSeek-Coder-V2-Instruct. The new model integrates the general and coding abilities of the two previous versions. For model details, please visit DeepSeek-V2 page for more information.

open-source ↓ 5.6K
DeepSeek-V3.2-Speciale logo
DeepSeek-V3.2-Speciale
deepseek-ai

DeepSeek-V3.2: Efficient Reasoning & Agentic AI

open-source ↓ 5.5K