LLM
Meituan Releases LongCat-2.0: A 1.6T-Parameter Open MoE Model with Native 1M Context and LongCat Sparse Attention
Meituan has released LongCat-2.0, a 1.6 trillion-parameter Mixture-of-Experts language model with native...
MarkTechPost
LLM · Other Companies
LLM
Alex Karp, frontier models and the real fight for Enterprise AI
Palantir Technologies CEO Alex Karp critiques frontier model vendors, arguing they aim to extract enterprise knowledge...
SiliconANGLE
LLM
Structured PDF-to-JSON: A Guide to Open-Source Extraction Models in 2026
The article discusses open-source PDF-to-JSON extraction models that convert unstructured enterprise data from PDFs...
MarkTechPost
LLM
UK's AI Security Institute finds standard benchmarks systematically underestimate what AI agents can actually do
The UK's AI Security Institute found that standard AI benchmarks systematically underestimate agent capabilities by...
The Decoder
LLM
GPT and Claude failed Bridgewater's finance tests because the right answers were never public
Bridgewater and Thinking Machines Lab have fine-tuned a Qwen3-235B model achieving 84.7% accuracy on financial tasks,...
The Decoder
LLM
Tesla caps employee AI spending at $200 per week
Tesla has implemented a cap of $200 per week on employee AI spending, according to an internal memo reported by The...
The Decoder
LLM
Using Lift to Turn Research PDFs into Structured JSON with Controlled, Schema-Guided Field-Level Evaluation
This tutorial demonstrates a complete PDF-to-JSON extraction workflow using Lift, a language model tool optimized for...
MarkTechPost
LLM
NVIDIA Releases Nemotron-Labs-TwoTower: an Open-Weight Diffusion Language Model Built on a Frozen Autoregressive Nemotron-3-Nano-30B-A3B Backbone
NVIDIA released Nemotron-Labs-TwoTower, an open-weight diffusion language model built on a frozen autoregressive...
MarkTechPost
LLM
Meituan's LongCat-2.0 shows China can train massive AI models without Nvidia
Meituan successfully trained a 1.6 trillion parameter AI model using only Chinese chips, demonstrating that China can...
The Decoder
LLM
How SMBs Can Take Advantage of Generative AI
Small and medium-sized businesses can leverage generative AI by improving their data quality and developing a...
AI Business
LLM
Advances in Natural Language Processing Are Changing Professional Networking
Natural language processing is reshaping professional networking platforms by enabling more personalized and relevant...
AI News
LLM
Samsung and SK Hynix plan $590 billion chip investment as AI demand sends memory prices soaring
Samsung and SK Hynix are investing $590 billion in chip manufacturing and packaging facilities to meet surging AI data...
The Decoder
LLM
China’s Z.ai claims it can match Mythos on cybersecurity
China's Zhipu AI released GLM-5.2, an open-weight model that researchers claim matches Anthropic's Mythos in...
The Verge AI
LLM
Coinbase joins the rush to Chinese AI models as Western labs face a pricing stress test
Coinbase CEO Brian Armstrong is switching the company to Chinese AI models like GLM 5.2 and Kimi 2.7 with an automated...
The Decoder
LLM
Sina's open model VibeThinker-3B aims to show reasoning compresses well but factual knowledge doesn't
Sina Weibo's VibeThinker-3B, a 3 billion parameter open model, achieves performance comparable to much larger models...
The Decoder
LLM
Liquid AI Ships LFM2.5-230M with llama.cpp, MLX, vLLM, SGLang, and ONNX Support for On-Device Inference
Liquid AI released LFM2.5-230M, a 230M-parameter open-weight model optimized for on-device inference, achieving 213...
MarkTechPost
LLM
ByteDance's "iLLaDA" is a diffusion language model that keeps up with Qwen2.5
ByteDance and Renmin University researchers released iLLaDA, an 8B diffusion-based language model that uses a different...
The Decoder
LLM
Building Supervised Fine-Tuning Data from NVIDIA Open-SWE-Traces: Trajectory Parsing, Patch Analysis, Token Budgets, and Tool-Use Metrics
A tutorial on building supervised fine-tuning datasets from NVIDIA's Open-SWE-Traces dataset for training agentic...
MarkTechPost
LLM
DeepReinforce Releases Ornith-1.0: An Open-Source Coding Model Family That Learns Its Own RL Scaffolds
DeepReinforce released Ornith-1.0, an open-source coding model family built on Gemma 4 and Qwen 3.5 that learns its own...
MarkTechPost