DeepSeek launches V4 Preview with open weights and a 1M-token context

Skip to main content # DeepSeek V4 Preview Release 🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length. 🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models. 🔹 DeepSeek-V4-Flash: 284B total / 13B active params. Your fast, efficient, and economical choice. Try it now at chat.deepseek.com via Expert Mode / Instant Mode. API is updated & available today! 📄 Tech Report: 🤗 Open Weights: ### DeepSeek-V4-Pro​ [...] 📄 Tech Report: 🤗 Open Weights: ### DeepSeek-V4-Pro​ 🔹 Enhanced Agentic Capabilities: Open-source SOTA in Agentic Coding benchmarks. 🔹 Rich World Knowledge: Leads all current open models, trailing only Gemini-3.1-Pro. 🔹 World-Class Reasoning: Beats all current open models in Math/STEM/Coding, rivaling top closed-source models. ### DeepSeek-V4-Flash​ 🔹 Reasoning capabilities closely approach V4-Pro. 🔹 Performs on par with V4-Pro on sim

🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length. 🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models. 🔹 DeepSeek-V4-Flash: 284B total / 13B active params.… pic.twitter.com/n1AgwMIymu [...] DeepSeek V4 release cover with logo and bold “What You Need to Know” headline on blue abstract background. # DeepSeek V4 Released: Everything You Need to Know On April 24, 2026, DeepSeek released preview versions of its long-awaited V4 series, an open-source Mixture-of-Experts family with two models and a native one-million-token context window. DeepSeek-V4-Pro has 1.6 trillion total parameters with 49 billion activated per token. DeepSeek-V4-Flash has 284 billion total parameters with 13 billion activated. Both are available today on Hugging Face with open weights, through the DeepSeek API, and via chat.deepseek.com. DeepSeek is one of the headline names in open sourc

DeepSeek @deepseek\_ai ·Follow 🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length. 🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models. 🔹 DeepSeek-V4-Flash: 284B total / 13B active params. Show more Image 45.3KReply Read 1.6K replies ## The Two Models DeepSeek split V4 into a Pro tier and a Flash tier, both Mixture-of-Experts. | Model | Total Params | Active Params | Positioning | --- --- | | DeepSeek-V4-Pro | 1.6T | 49B | Frontier. Rivals top closed-source models | | DeepSeek-V4-Flash | 284B | 13B | Fast, efficient, economical | [...] Image DeepSeek DeepSeek @deepseek\_ai 🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length. 🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models. 🔹 DeepSeek-V4-Flash: 284B total / 13B active params. Imag

| Model | Parameters | Activation | Pre-training Data | Context Length | Open Source | API Service | Web/App Access Mode | --- --- --- --- | | deepseek-v4-pro | 1.6T | 49B | 33T | 1M | ✓ | ✓ | Expert Mode | | deepseek-v4-flash | 284B | 13B | 32T | 1M | ✓ | ✓ | Fast Mode | Starting today, you can chat with DeepSeek-V4 at chat.deepseek.com or via the official app. The API is also live — simply set model\_name to deepseek-v4-pro or deepseek-v4-flash to get started. We have reported the speculation and pre-release analysis earlier (see our DeepSeek V4 expectation guide and technical deep dive), we now have confirmed, official details straight from the source. The following covers exactly what shipped, what's new, and what it means if you're building with or evaluating AI models today. [...] ## DeepSeek-V4-Pro: Rivaling Top Closed-Source Models Significantly enhanced agentic capabilities. Compared to its predecessor, DeepSeek-V4-Pro shows a dramatic improvement in agent tasks. In Age

Looks like the whale is back again to take back its crown, but it may take a bit longer for it to do that. Cuz I'll be honest, this model isn't what I expected. The DeepSeek team today is back and they are releasing two new models as a part of the V4 release. This is where I'd like to introduce DeepSeek version 4 preview. This is where they are open-sourcing two new models, the Pro and Flash, built around a cost-effective 1 million context length. The DeepSeek version 4 Pro is the flagship model with a 1.6 trillion total parameter size, and it has 49 billion active parameters, claiming that it is the top open-source performative model, better than every open-source model in reasoning, STEM, coding, agentic workflows, and world knowledge, as well as rivaling leading closed-source models. [...] as well as rivaling leading closed-source models. But I personally see that as benchmark maxed, meaning that the team optimized heavily for benchmarks in their favor. From my testing, I still see

2.4M user avatar DeepSeek @deepseek\_ai Apr 25 🔥DeepSeek-V4-Pro API is 75% OFF until May 5th, 2026, 15:59 (UTC Time)! Don't miss out on this massive discount. 🛠️Integration Updates: 🔹Claude Code: Set model to deepseek-v4-pro[1m] to unlock 1M context! 🔹OpenCode: Update to v1.14.24+ 🔹OpenClaw: Update to v2026.4.24+ Check 1.3M user avatar DeepSeek @deepseek\_ai Apr 24 🚀 DeepSeek-V4 Preview is officially live & open-sourced! Welcome to the era of cost-effective 1M context length. 🔹 DeepSeek-V4-Pro: 1.6T total / 49B active params. Performance rivaling the world's top closed-source models. 🔹 DeepSeek-V4-Flash: 284B total / 13B active params. # See @deepseek\_ai's full profile Sign up Log in