Thinking Machines amps up its bet against one-size-fits-all AI with its first open model, Inkling
It's the company's first public proof point after a year and a half spent building AI infrastructure largely out of public view.
Get smart on it
An AI startup founded by a former OpenAI executive released its first in-house model as open-weight software, meaning outside developers can download and modify it directly, unlike the proprietary models from major AI labs. The company is betting that customizable AI tailored to specific organizations will outperform one-size-fits-all models, and is positioning its model as a starting point for companies to fine-tune through its customization platform rather than as a finished product. This approach contrasts with how OpenAI, Anthropic, and Google market their models as general-purpose chatbots, and aligns with recent arguments from major tech leaders that enterprises overpay for proprietary AI while also surrendering valuable business knowledge. The startup reached this milestone in approximately nine months, significantly faster than its larger competitors, though it remains unclear how the company plans to generate revenue since anyone can download and run the open-weight model without paying.
Related news
NVIDIA AI Releases Nemotron 3 Embed: An Open Embedding Collection Whose 8B Checkpoint Ranks #1 on RTEB
Embedding models decide which passages an agent ever sees. NVIDIA released Nemotron 3 Embed model to work on that layer. It targets production-scale RAG, agentic retrieval, code retrieval, and agent memory. What is Nemotron 3 Embed? The model collection includes three open checkpoints. Nemotron-3-Embed-8B-BF16 is the accuracy-first option. Nemotron-3-Embed-1B-BF16 carries the same design into a smaller footprint. Nemotron-3-Embed-1B-NVFP4 is the Blackwell-optimized 4-bit path. All thre
Moonshot AI Releases Kimi K3: A 2.8 Trillion Parameter Open MoE Model With Kimi Delta Attention and 1M Context
Moonshot AI just released Kimi K3. It is a 2.8-trillion-parameter model with native vision and a 1-million-token context window. Moonshot calls it the world’s first open 3T-class model. What is Kimi K3? Kimi K3 is a sparse Mixture-of-Experts (MoE) model built on two architectural updates. Those are Kimi Delta Attention (KDA) and Attention Residuals (AttnRes). Both change how information flows across sequence length and model depth. K3 targets long-horizon coding, knowledge work, an
