Models | There's An AI For That

By DeepSeek

DeepSeek-V4-Flash is DeepSeek’s faster, smaller, and more economical V4 model for efficient large-scale use. It has 284B total parameters with 13B active parameters, supports a 1M token context window, and is positioned as a high-speed, cost-effective model whose reasoning comes close to V4-Pro.

📚Large Language Models 💬Chatting 💻Conversational coding 🧠AI inference

NewMultimodal

Released 1mo ago

Gen 3 DeepSeek

Deepseek V4 Pro

By DeepSeek

DeepSeek-V4-Pro is DeepSeek’s new flagship open-source model for top-end reasoning, coding, world knowledge, and agentic work. It uses a MoE architecture with 1.6T total parameters and 49B active parameters, supports a 1M token context window, and is positioned as rivaling leading closed-source models while leading current open models on several major capabilities.

📚Large Language Models 💬Chatting 💻Conversational coding 🧠AI inference

NewMultimodal

Released 1mo ago

Gen 7 DeepSeek

DeepSeek OCR 2

By DeepSeek

Second-generation DeepSeek OCR model, “Visual Causal Flow,” aimed at more human-like visual encoding, with dynamic resolution support and strong document-to-Markdown and layout-aware OCR for images and PDFs.

📜OCR

Text

Released 4mo ago

Gen 7 DeepSeek

DeepSeek 3.2 Speciale

By DeepSeek

DeepSeek-V3.2-Speciale is a 685B-parameter research-only variant of DeepSeek-V3.2 that pushes open-weight reasoning ability to the limit, but disables tool calling and is intended purely for experimentation rather than everyday agent use.

🔍Research 📊Data analysis 🔍SEO strategies 📄PDF Management

Text

Released 6mo ago

Gen 7 DeepSeek

Deepseek 3.2

By DeepSeek

DeepSeek V3.2 is the core, stable 3.2-series model designed as a strong general-purpose LLM. It builds on the DeepSeek V3 architecture and delivers balanced performance across reasoning, writing, coding, and multilingual tasks.

📝Writing 💻Coding 📄Resumes 📚Stories

Text

Released 6mo ago

Gen 7 DeepSeek

DeepSeekMath V2

By DeepSeek

DeepSeek-Math-V2 is a math-specialized LLM built on DeepSeek-V3.2-Exp-Base, trained to generate and verify step-by-step proofs. It uses a learned verifier as a reward model so the generator learns to fix its own reasoning, reaching gold-level scores on contests like IMO 2025, CMO 2024, and near-perfect Putnam 2024 with scaled test-time compute.

➕Math lessons 🔍SEO content 🌐Websites ➗Math problems

Text

Released 6mo ago

Gen 7 DeepSeek

DeepSeek OCR

By DeepSeek

LLM-centric OCR model using “Contexts Optical Compression” to explore visual-text compression and provide fast streaming and batch OCR for images and PDFs via vLLM and Transformers runtimes.

📜OCR

Text

Released 7mo ago

Gen 7 DeepSeek

DeepSeek V3.2 Exp

By DeepSeek

DeepSeek v3.2 Exp is an experimental build of the DeepSeek V3 line, tuned for deeper reasoning and stronger coding while keeping latency practical. It supports long context, function/tool calling, and schema-true JSON—great for RAG, agents, and repo-scale tasks when you want extra accuracy.

📚Academic research 🧑‍🏫Design mentoring 🌟Life advice 🔍Mep estimations

Text

Released 8mo ago

Gen 7 DeepSeek

DeepSeek V3.1 Terminus

By DeepSeek

DeepSeek-V3.1-Terminus is DeepSeek’s flagship reasoning model, tuned for difficult analysis, math, and coding. It supports very long context, function/tool calling, reliable JSON outputs, and an optional extended-thinking mode—ideal for enterprise RAG, agents, and high-stakes workflows.

📊Data analysis 💻Coding 📚Summaries

Text

Released 8mo ago

Gen 7 DeepSeek

DeepSeek-R1

By DeepSeek

DeepSeek R1 is a reasoning-first large language model built to solve complex problems with explicit multi-step thinking. It excels at math, coding, and logical analysis, supports long context, tool/function calling, and structured JSON outputs, and can trade latency for higher accuracy via extended "thinking" budgets.

💻Coding 📊Data analysis 🐞Debugging 🇯🇵Japanese lessons

Text

Released 1y ago

Gen 6 DeepSeek

DeepSeek V3

By DeepSeek

Text

Released 1y ago

Gen 6 DeepSeek

DeepSeek v2.5

By DeepSeek

Text