Latest stories
DeepSeek V4 Launches with 1.6T Parameters, Claims 10-50x Cheaper Pricing
DeepSeek unveiled its V4 model family on April 24, 2026, featuring two text-only variants with up to 1.6 trillion parameters and a...
DeepSeek Retires Legacy API Aliases: Price Changes and Migration Guide
DeepSeek retired the legacy API aliases 'deepseek-chat' and 'deepseek-reasoner' on July 24, 2026, at 15:59 UTC. Developers must up...
DeepSeek-V4-Flash Goes Official: Agent Benchmarks Surpass V4-Pro-Preview
DeepSeek has launched the official deepseek-v4-flash API in public beta, featuring improved agent capabilities that outperform V4-...
DeepSeek vs Stepfun: Two Chinese AI Labs, Two Divergent Strategies for 2026
DeepSeek focuses on reasoning and low-cost APIs, while Stepfun bets on multimodal generation including video and audio. Here's how...
DeepSeek Retires R1, Replaces with V4: A New Era for Open-Source Reasoning
DeepSeek has officially retired its R1 reasoning model, replacing it with V4. The deepseek-reasoner alias was decommissioned on Ju...
DeepSeek R2 Early Release Faces Technical and User Experience Hurdles
DeepSeek's accelerated release of its R2 AI model aims to capture market leadership but exposes the system to stability issues, in...
Inside DeepSeek's DSpark: How Speculative Decoding Speeds Up LLM Inference Without Losing Quality
DeepSeek introduces DSpark, a speculative decoding technique that accelerates large language model inference while maintaining out...
DeepSeek Raises $7.4B at $50B+ Valuation with Unprecedented Founder-Friendly Terms
DeepSeek secures $7.4 billion (50 billion yuan) at a valuation exceeding $50 billion, featuring a highly founder-centric limited p...
DeepSeek AI: China's Open-Source Search Revolution Unveiled
DeepSeek, an AI-driven search technology from MIT's AI Lab, promises deeper, context-aware results beyond traditional keyword sear...
DeepSeek V4 Unveiled: 1.6T MoE Model with 1M Context and Agent Integrations
DeepSeek releases V4, a 1.6 trillion parameter MoE model with 49B active parameters, 1M context via DSA + Hybrid Attention, Muon o...
Claude Opus 4.8 Launches: Benchmark Gains and 3x Cheaper Fast Mode vs DeepSeek V4 Pro
Anthropic released Claude Opus 4.8 with improved benchmarks, dynamic workflows in Claude Code, and a 3x cheaper fast mode. A head-...
DeepSeek V4: 4 Chat Modes, Chaining, 1M Context – Pro Tips
DeepSeek's latest guide reveals how to master V4's four chat modes, chain workflows, combine file uploads with web search, and lev...