Home
Tools
Models
App
Prompts
News
Topics
Submit for inclusion
English
English 日本語 한국어 Português español Deutsch Русский Français 繁體中文 简体中文
option
Home Close
Home
Tools
Category Ranking Generate Cover Generate Image Skill
Models
Large Language Model Multimodal Model
App
Prompts
Prompts Image Prompts
News
News Flash News
Topics
Submit for inclusion
Submit App Submit Tool
Create an account Sign In
Change Language English Change Language
Settings
English EN 日本語 JA 한국어 KO Português PT español ES Deutsch DE Русский RU Français FR 繁體中文 ZH-TW 简体中文 ZH-CN
Home
Flash News
Content
JerryGonzalez
JerryGonzalez
April 24, 2026

DeepSeek launches its new-generation open-source large model DeepSeek-V4, featuring a 1 million token context window. The model comes in two versions: a lightweight Flash edition and a high-performance Pro version. DeepSeek-V4-Pro excels in mathematics, STEM, coding, and agent capabilities, rivaling top closed-source models. It is offered at a disruptive price point, with input costs as low as 1 yuan per million tokens for the Flash version.

DeepSeek launches its new-generation open-source large model DeepSeek-V4, featuring a 1 million token context window. The model comes in two versions: a lightweight Flash edition and a high-performance Pro version. DeepSeek-V4-Pro excels in mathematics, STEM, coding, and agent capabilities, rivaling top closed-source models. It is offered at a disruptive price point, with input costs as low as 1 yuan per million tokens for the Flash version.
0
0
Share Click to copy the link
Comments (0)
0/300
Submit
Author JamesBaker JamesBaker September 15, 2026
StepXingchen launched StepAudio3 series on Sept 15, featuring five models: StepAudio3Realtime, ASR, TTS, Gen, and Music. Several models ranked first globally on Artificial Analysis. StepAudio3Realtime achieves 98.9% conversational dynamics score with full-duplex interaction. StepAudio3ASR ties for first globally with 1.7% word error rate. StepAudio3TTS offers realistic voice generation with emotional adjustment. StepAudio3Gen enables comprehensive audio content creation from natural language. StepAudio3Music supports music creation via lyrics or ABC notation, handling harmony and arrangement automatically. All models are available on the StepXingchen open platform for diverse applications.
0
0
Author JerryGonzález JerryGonzález September 15, 2026
Canadian AI researcher David Krueger calls for an immediate indefinite pause on cutting-edge AI development, arguing that current industry proposals for slowing progress are insufficient. He criticizes existing testing methods as unreliable, noting that AI can bypass evaluations. Krueger warns that merely slowing development is inadequate to prevent existential risks, urging stronger and faster regulatory action to mitigate the threat of human extinction from advanced artificial intelligence.
0
0
Author RalphGarcia RalphGarcia September 15, 2026
Anthropic launches Claude Opus 5.2 with faster response and autonomous iteration. Apple releases iOS 27, upgrading Siri AI and optimizing older devices. GPT-6 Astra outperforms rivals in a vending machine simulation. Ant Group open-sources SingProbe, offering real-time risk blocking with under 0.5% overhead. Instagram restricts AI photo editing behind a subscription. Shanghai AI Lab releases ShuSheng-S2, a multimodal model excelling in scientific tasks. WeChat clarifies Xiao Wei AI privacy policies. StepX unveils StepAudio3 series, featuring advanced voice and music generation capabilities.
0
0
Author RobertHarris RobertHarris September 15, 2026
StepZenith launched the StepAudio3 series, featuring five models covering real-time interaction, speech recognition, text-to-speech, audio generation, and music creation. StepAudio3Realtime ranks first globally in conversational dynamics and voice reasoning accuracy. StepAudio3ASR achieves a 1.7% word error rate, tied for first worldwide. The suite supports multi-language, dialects, and professional domains like medical and legal. StepAudio3Gen and StepAudio3Music enable comprehensive audio content production with natural language control over style, melody, and rhythm. All models are available on the StepZenith Open Platform, with multiple rankings leading globally in Artificial Analysis.
0
0
/live/7340
Author AnthonyMartinez AnthonyMartinez September 15, 2026
Major firms like NVIDIA, Palantir, and Booz Allen Hamilton are restricting Anthropic and OpenAI models in sensitive projects due to data retention concerns. NVIDIA limits Claude for low-sensitivity tasks while using its Nemotron model for confidential work. Booz Allen demands stricter contracts, C Spire bans data training, and Novo Nordisk prohibits proprietary data input. This trust crisis drives demand for private cloud environments, with Microsoft capitalizing on the shift toward enhanced data security in enterprise AI.
0
0
OR