Skip to contentDEEKOYA AI CoursesFLAT 20% OFFEnroll Now
Deekoya AI

AI News

17,651 stories from 2891 sources · 1115 in the last 24 hours · refreshed hourly. Every story links to the original publisher.

Today's AI news from trusted publishers and official AI lab blogs, refreshed about every 30 minutes.

News today: 1,834Last updated:
  1. Google Open-Sources Harness2, a Recursive Agent Harnessing FrameworkMarkTechPost · modelsGoogle Research released Harness2, an Apache-2.0 framework that rewrites an agent's harness at test time. One refinement step lifted GLM-5 by 11.5 points on WorkBuddy-Bench with no weight updates. The post Google Open-Sources Harness2, a Recursive Agent Harnessing Framework appea
  2. How to Make Zero-Training Tabular Predictions with NVIDIA Structured Data ModelsMarkTechPost · hardwareIn this tutorial, we implement Structured Data Models, NVIDIA s new library that puts several pretrained foundation models for tables behind one PyTorch interface, and use it to make predictions without training. The library wraps TabICLv2 and KumoTabular for single tables and Ku
  3. Meet Voxel51 3.0: A Unified Multimodal Data Platform for Physical AIMarkTechPost · toolsVoxel51 3.0 merges search, curation, annotation and evaluation of multimodal sensor data into one platform for physical AI teams. It adds Voxel51 Agent with 45+ MCP tools, dataset versioning, and a free self-service tier. The post Meet Voxel51 3.0: A Unified Multimodal Data Platf
  4. Meet Kyutai REKI: The 2.4B VLA Model Bringing Efficient Memory to Robot ManipulationMarkTechPost · modelsKyutai, with Inria and École des Ponts, has released REKI, a 2.4B vision-language-action model (VLA) for robot manipulation. The REKI vision-language-action model is pretrained only on public data. It also carries a cheap memory of its recent past. On three out-of-distribution be
  5. DeepSeek Just Released V4.1-Flash-1017, Scores 77.0 on DeepSWE, Opens Search APIMarkTechPost · modelsDeepSeek has just released DeepSeek-V4.1-Flash-1017, a new checkpoint of its V4.1-Flash model. It is live on the DeepSeek website, app and API. DeepSeek claims stronger scores, faster task completion and shorter outputs than the September V4.1-Flash release. The company calls it
  6. 20 Jev Alternatives: The Decision Models Taking On TypeSafe in 2026MarkTechPost · modelsTypeSafe launched Jev on September 15, 2026. It is a decision model, not a chatbot. You define the allowed answers, and Jev returns a label, a score or a yes/no probability. In about 4 weeks, at least 20 Jev alternatives shipped. They come from Cloudflare, OpenAI, Perplexity, Liq
  7. Perplexity Open-Sources SPACE, a Firecracker MicroVM Sandbox for AI AgentsMarkTechPost · general
  8. Datalab Releases OmniParseBench: An Open OCR Benchmark With 16,288 Pass/Fail Tests to Audit OCR ParsersMarkTechPost · modelsDatalab released OmniParseBench, an open OCR benchmark with 16,288 yes/no tests on 2,937 pages in 92 languages. Every score is a pass rate you can slice by tag, script or language. The post Datalab Releases OmniParseBench: An Open OCR Benchmark With 16,288 Pass/Fail Tests to Audi
  9. Fish Audio Releases Drama 3, a 35B Open-Weight MoE Text-to-Speech ModelMarkTechPost · modelsFish Audio released Drama 3, a 35B mixture-of-experts text-to-speech model with about 3B active parameters. It takes plain-language direction, covers 83 languages and ships open weights under a revenue-capped community license. The post Fish Audio Releases Drama 3, a 35B Open-Wei
  10. 9 Best AI Gateways in 2026: Bifrost, Portkey, LiteLLM, Kong and 5 More RankedMarkTechPost · generalMarktechpost ranks the 9 best AI gateways for 2026. We compare Bifrost, Portkey, LiteLLM, Kong, TrueFoundry, Cloudflare, Vercel, OpenRouter and Orq.ai on latency, pricing, licensing, governance and deployment. The post 9 Best AI Gateways in 2026: Bifrost, Portkey, LiteLLM, Kong a
  11. Liquid AI Releases LFM2.5-24B-A2B: A 2B-Active MoE for On-Device AgentsMarkTechPost · modelsLiquid AI released LFM2.5-24B-A2B, an open-weight 24B Mixture-of-Experts reasoning model with 2B active parameters. It targets local agents, scoring 48.09 on LiveClaw and decoding 184 tokens/s on an Apple M5 Max. The post Liquid AI Releases LFM2.5-24B-A2B: A 2B-Active MoE for On-
  12. Google Releases Gemini 4 Argon API With 1M Token Output LimitMarkTechPost · modelsGoogle has made Gemini 4 Argon generally available on the Gemini API, AI Studio and the Gemini app. The sparse MoE model writes up to 1M output tokens and leads the Vals Index at 68.90%. The post Google Releases Gemini 4 Argon API With 1M Token Output Limit appeared first on Mark
  13. Meta Launches Muse Wonder and Muse Wonder Full: Benchmarks, Pricing and How to Use ItMarkTechPost · modelsMeta released Muse Wonder and Muse Wonder Full, 1M-context agent models. Muse Wonder costs $4/$20 per million tokens on Meta Model API. Wonder Full, on Muse Code plans, leads DeepSWE and SWE-Marathon. The post Meta Launches Muse Wonder and Muse Wonder Full: Benchmarks, Pricing an
  14. How to Use Meta s Rebalancer for Task Placement OptimizationMarkTechPost · generalIn this comprehensive coding tutorial, we explore Meta's Rebalancer assignment solver using its C++ core and Python bindings. Learn how to build datacenters, manage task placements across racks, handle constraints and goal priorities, and compare local search against the exact Hi
  15. Z.ai GLM-Legal Beats Claude Fable 5 and Kimi K3 on Harvey LABMarkTechPost · modelsZ.ai’s GLM-Legal is a legal AI model built on GLM-5.3-Flash. It scores 19.2% all-pass on Harvey’s Legal Agent Benchmark, nearly 3x its base, at $0.15 per test. Access is by application. The post Z.ai GLM-Legal Beats Claude Fable 5 and Kimi K3 on Harvey LAB appeared first on MarkT
  16. Alibaba Qwen Introduces QwenGyre: Elastic Reinforcement Learning RL for XLong-Horizon AgentsMarkTechPost · modelsAlibaba s Qwen team has introduced QwenGyre, an elastic reinforcement learning RL framework for extreme-long-horizon agent training. QwenGyre is an online reinforcement learning system from Alibaba that trains coding agents through unmodified, black-box harnesses. It targets roll
  17. Top 12 AI Live Translation Tools in 2026: Latency, Languages, PricingMarkTechPost · generalWe checked 12 AI live translation tools, including Palabra.ai, Google Live Translate, DeepL Voice, Teams Interpreter and Wordly, against official pages for languages, latency, platforms and pricing. The post Top 12 AI Live Translation Tools in 2026: Latency, Languages, Pricing ap
  18. Perplexity AI Releases pplx-embed-v2-late: A 0.6B Edge Model and a 9B Model Scoring 92.4% on MADQAMarkTechPost · models
  19. Anthropic Releases Claude Haiku 5.5: A Small Model With 1M Context Priced at $0.10 per Million Input TokensMarkTechPost · models
  20. Mistral AI Releases Mistral Large 4 (Le Chonk): A 1.05T Parameter Multimodal MoE ModelMarkTechPost · models
  21. Perplexity Releases pplx-embed-v2-context-9b-preview: A Contextual Embedding Model That Retrieves Answers and Their Supporting EvidenceMarkTechPost · models
  22. Perplexity Introduces Photon: A Rust-Based Retrieval Engine That Cuts p99 Latency From 800 ms to 65 msMarkTechPost · general
  23. Anthropic Releases Claude Sonnet 5.5: 70.6% on Terminal-Bench 4.0 at the Same $2/$10 PriceMarkTechPost · models
  24. Google Research Introduces an AI Video Co-Director: 4 Agentic Frameworks for Coherent, Minutes-Long Video GenerationMarkTechPost · research
  25. Sarvam AI Releases Saaras V4: A Speech-to-Text Model for All 22 Indian Languages and Global EnglishMarkTechPost · models · 🇮🇳
  26. Perplexity Trains Its Computer Agent on Real Mistakes With Hint-Guided Self-DistillationMarkTechPost · tools
  27. SpaceXAI Releases Grok 4.7: A Larger Base Model at the Same $2/$6 Price as Grok 4.6MarkTechPost · models
Load more

Frequently asked questions

How often is AI news updated?

About every 30 minutes from public RSS feeds.

Do you write the news?

No. Headlines link to the original publisher.