Open-Source AI
Open models — Llama, Mistral, DeepSeek, Qwen and others — are catching up to closed labs and let you run AI on your own hardware, vendor-free. That shifts the balance of power, the safety debate and the economics of the field. Here is everything on open-source AI: open-weight releases, licenses, local deployment and the race against closed models.
Thinky Releases Open Inkling 975B-A41B Model Under Apache 2.0 License
Thinky has released Inkling, a 975B-parameter multimodal MoE model with open weights under Apache 2.0, plus a compact 276B version, Inkling-Small.
Evidence Chain Evaluation: An AI Fact-Checker That Is Allowed to Say "Not Sure" Instead of Issuing False Verdicts
Researchers unveiled Evidence Chain Evaluation, a fact-checking framework that abstains when evidence is weak and holds 97.8% accuracy on the claims it answers.
Kimi K3: The Largest Open AI Model Yet, Claimed to Match Opus 4.8 at Sonnet 5 Pricing
Moonshot AI has released Kimi K3, a 2.8-trillion-parameter MoE model (about 50B active) and the largest open model ever. It claims Opus 4.8-level quality at Sonnet…
Small Qwen Language Models Ignore Instructions but Look Accurate on Benchmarks
CRA-Bench and CRA-Net: Catching Harm That Builds Up Over the Course of an LLM Conversation
Spectral-LSH: Up to 16× Prompt Compression Without Fine-Tuning to Speed Up LLM Inference
DeepSeek Raised Its Registered Capital 65% in 2026, and Its Founder Says AGI Is the Only Goal
LISA: Sparse Attention Speeds Up Long LLM Reasoning by 50% Without Retraining
The LISA method replaces the standard attention mechanism in language models and speeds up long reasoning by 50% at a 16K-token context, with no training from…
Confidential Inference on NVIDIA H100 with Intel TDX: Privacy Costs 17–21% of Speed
Researchers measured the cost of confidential computing for LLM inference: on an NVIDIA H100 under Intel TDX, secure mode cuts throughput by 17–21% and slows time…
Tencent Hunyuan's Multimodal AI Chief Hu Han Departs as Team Pivots to World Models
Open vs. Closed: GLM-5.2 and DeepSeek V4-Pro Now Trail the Frontier by Just 4–7 Months
DeepSeek's $7.4B Round Pushes Asian Startup Funding to a Three-Year High
Sam Altman's 2022 Letter: OpenAI Considered Open Source to Choke Off Competitors
Moonshot AI Halts Kimi K3 Subscriptions as Demand Outstrips Compute Capacity
Beijing-based Moonshot AI has stopped accepting new Kimi K3 subscribers after demand nearly exhausted the company's compute capacity within 48 hours.
Moonshot Releases Kimi K3, the Largest Open Model Yet at 2.8 Trillion Parameters
Moonshot AI unveiled Kimi K3, a 2.8T-parameter model and the largest open model to date. It beats DeepSeek V4 Pro on size but carries a record price for a Chinese…