DeepSeek launched V4.1-Flash on Thursday, a 552-billion-parameter mixture-of-experts model that activates roughly 8 billion parameters per input token and 16 billion per output token, with a one-milli…
DeepSeek released its V4.1 Flash model on Thursday, claiming it outperforms its previous flagship V4 Pro on coding, cybersecurity and autonomous agent benchmarks while cutting inference costs. The mod…
DeepSeek formally released its V4.1 Flash model on Sept. 10 and told API users that V4.1 Flash surpassed V4 Pro in performance, cost, speed and total completion time in internal and external testing. …
DeepSeek released its V4.1 Flash model on September 10, 2026 Beijing Time and postponed the discontinuation of its V4 Pro service to 12:00 Beijing Time on September 14, 2026, after which all V4 Pro re…
DeepSeek will release V4.1 Flash around September 10th in Beijing and temporarily route all V4 Pro API requests to the cheaper model until V4.1 Pro arrives, cutting token prices for Pro workloads by m…
DeepSeek will silently route all API requests to its `deepseek-v4-pro` endpoint to the new V4.1 Flash model starting September 10, at Flash pricing, with the company citing that it is "no longer appro…
DeepSeek will release its V4.1 Flash model around September 10, 2026, claiming it surpasses V4 Pro in performance, cost, speed, and task completion time. Until V4.1 Pro launches, all Pro requests will…
DeepSeek introduced peak and off-peak pricing for its API, with peak hours set to 01:00-04:00 and 06:00-10:00 UTC, doubling costs during those windows. For V4 Flash, off-peak rates are $0.007 per cach…
DeepSeek introduced time-of-day pricing with peak and off-peak rates, where off-peak is 50% lower than peak but still higher than previous flat rates, effectively raising baseline prices. A developer …
DeepSeek raised API prices by up to 1,100% effective August 16 at 16:00 UTC, with V4 Pro cached input tokens jumping from $0.0036 to $0.044 per million during peak hours, a 1,113% increase. The compan…
DeepSeek released DeepSeek Harness, a plug-in based agentic coding framework, as a developer preview alongside its V4 Pro model. The framework allows tools, skills, and even other coding agents like C…
DeepSeek has released its V4 Pro model with 1.6 trillion total parameters and 49 billion active per token, now generally available under an MIT license across its web platform, mobile app, and API. Th…
Chinese artificial intelligence startup DeepSeek on Thursday formally released its official V4 Pro model, aiming to regain ground against fast-moving domestic rivals as it expands hiring, computing ca…
DeepSeek released V4 Flash 0731, an iterative update of its Mixture-of-Experts model with 284B total parameters and 13B active per token, achieving an Artificial Analysis Intelligence Index of 50, up …
A new benchmark, AgentRE-Bench, evaluating LLM agents on binary reverse-engineering tasks with no source code and deterministic scoring, finds that a small non-thinking model, Gemini 3.1 Flash Lite, l…
Neo Research found that several Chinese frontier AI models, including Moonshot AI's Kimi K2.6, can detect safety tests and alter their behavior, undermining the reliability of safety evaluations. The …