2026-10-14

Gguf's: Free Demo: "LFM2.5-24B-A2B is a 24B-parameter reasoning model with only 2B active parameters, built for fast, private agents on-device . It is the largest and final model in the LFM2.5 family, which now spans two orders of magnitude: from LFM2.5-230M to LFM2.5-24B-A2B. Best-in-class agent : Competitive with Qwen3.6-35B-A3B, a model with ~50% more total and active parameters, on agentic and tool-use tasks, scoring 48.09 on LiveClaw versus 38.19. Reasoning at MoE speed : Always thinks befo

Release: llm-meta-ai 0.2 Meta AI release Muse Wonder (and Muse Wonder Full), their new "most capable models". These look like they may be Meta's first Fable class models - Artificial Analysis scores Muse Wonder (Max) as equivalent to GPT-6.1 Sol. I updated my llm-meta-ai plugin for the new model. Here are pelicans for Muse Wonder for the six reasoning levels: minimal, low, medium, high, xhigh, max. This is the max one: For comparison, here are pelicans for Muse 1.3 (released September 2nd) and M

Research: Profiling web-edit with Python 3.15: flame graphs and source heatmaps I put the new Python 3.15 profiler through its paces by having GPT-6 Astra Extra High in Codex run it against my new web-edit app : I need a Python 3.15 flame graph and heatmap demo of the "web-edit" app from PyPI - research first then make me those demos Here's the resulting example flame graph and heatmap . Tags: profiling , python , codex

Topgolf and VMware "ran deep" before the migration.

AI leaders joined whistleblowers and security leaders at a sweeping parliamentary session, while a few hundred meters away Britain’s spy chief warned AI could be used to create new bioweapons.

Z.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 MarkTechPost .

Power sampling improves large language model (LLM) reasoning at test time by sampling from a sharpened version of the base model's distribution. Its sampler, PowerMH, is slow because its Metropolis--Hastings (MH) updates are sequential: each accept/reject decision sets the state from which the next proposal is generated. Prior work reduces this cost by changing the sampler or the model (e.g., sequential Monte Carlo, approximate lookahead, distillation). We propose Predictive SubTree Prefetching

As text-to-SQL systems move beyond demonstrations toward real-world deployment, ambiguity in user questions becomes a primary source of errors. Such ambiguities are often subtle, domain- or data-specific, and can silently cause system outputs to deviate from the user's true intent. Ambiguity is traditionally addressed through conversational clarification, which is often inefficient, cognitively demanding, and poorly aligned with real-world user workflows. We propose structured disambiguation, a

Large language models have achieved remarkable advances in complex reasoning. However, scaling reasoning post-training with ever-larger corpora can substantially increase training cost without proportional performance gains. In this study, we therefore ask: how far can reasoning supervision be compressed without sacrificing capability? To this end, we propose an uncertainty-guided progressive contraction framework that concentrates supervision and optimization on the reasoning trajectories most

I'm excited to see our Lemonade embedded SDK getting used out there!

Hi HN, I'm Niz. Pulse Local is a budgeting app for Mac (Apple Silicon, beta) that answers questions about your spending with models that run on your machine. Local mode needs no account. The ledger and your questions stay on the machine, and the app sends no analytics or crash logs. I felt there are not enough consumer apps that run locally with no calls to the outside world that help us manage sensitive corners of our lives. This app in its local variant only downloads a couple of models and oc

It's not clear exactly for what purpose the recording device was used, but FBI director Kash Patel said the scheme was to "capture sensitive communications about the company’s super intelligence technology."

The company asked its AI models to figure out what the device was.

When Keelan Miskell began winding down his crypto startup this year, he didn’t think its data was worth much: the users were anonymous, the company just a few years old. But he’s since turned the firm’s digital archives into real money. He’s netted roughly $25,000 for his code base and a still-pending amount (likely around…

Higgsfield brings creators longer videos with GPT-6 Astra, precise visuals with GPT-Image-2.5, and a path to voice filmmaking with GPT-Live.

MIT Technology Review Explains: Let our writers untangle the complex, messy world of science and technology to help you understand what’s coming next. You can read more from the series here. Late last month, Anthropic released a report arguing that the model GLM-5.3 poses a major cyberattack risk. The report’s core concern is that the…

GitHub: Docs: Discord: Hello, Yesterday was the official release of Termix official + community plugins. With the changes in this and last week’s update, all primary functions of Termix have been moved to plugins, and a small UI redesign came along with it. If you’re not familiar, Termix is a free, open source, self-hosted server management tool that is now built on plugins. The core keeps your hosts, credentials, users, sharing, sessions, etc. Everything else, like the SSH terminal, remote desk

Didn't realize the V100 were still so fast, anyone have first hand experience on these?

Chatbot prompts reveal the engineer’s alleged efforts to set up a covert listening device days before one was found in a meeting room used by high-level executives.

An OpenAI engineer was arrested after the company said it discovered a "sophisticated recording device" hidden in one of its offices. The employee was identified as Abdulrahman Alfozan, a 35-year-old Saudi national. Alfozan is charged with using or endeavoring to use a device to intercept oral communications, according to a press release from the Department […]

About 11 months ago i revived the amazing but unfortunately unmaintained spotify-tui. Since then I have been working on it almost non stop, finding a lot of problems on the way. One of the main issues I ran into were the Spotify web API rate limits. Third party clients like spotatui are dependent on a developer app made in the Spotify developer dashboard, which get rate limited a lot and Spotify has slowly been deprecating endpoints from it. The workaround for this used to be using a shared clie

Days after Ukrainian military drones took down several Russian data centers, President Volodymyr Zelenskyy claimed that the campaign had hurt Russia's disinformation efforts. "It suddenly became a little easier to breathe on many social platforms," Zelenskyy said Monday. "There was less hate, less division, and fewer provocations." Zelenskyy isn't the first to assert that the […]

Article URL: Comments URL: Points: 83 # Comments: 35

Article URL: Comments URL: Points: 39 # Comments: 40

University of Michigan spinout Laplace Intelligence uses AI agents and interactive simulation to stress-test robots and self-driving vehicles before deployment. The profitable startup is competing in Startup Battlefield at TechCrunch Disrupt, October 13-15 in San Francisco.

I've been reading about all these modders using AI to reverse-engineer games and software. I've been trying out photocraft and gridcraft and have just watched a GTA5 mod featuring Sonic. It's pretty damn impressive. As a struggling AMD Radeon owner I'd love to see AI generated open source CUDA implementation. Not only do I dislike NVIDIA, but there's delicious irony in using CUDA cores to do it. Anyone up for the challenge?

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Open models have real benefits. Do they outweigh the risks? Anthropic recently released a report arguing that a powerful new model released by the Chinese company Z.ai, poses a major cyberattack…

When solving complex spatial reasoning tasks, ordinary humans naturally revisit the scenes from task-relevant viewpoints and mentally reconstruct the spatial layouts from those viewpoints. But can Large Vision-Language Models (LVLMs) proactively think in the same way? We term this ability Active Spatial Mentalization (ASM), an intrinsic component of visual intelligence that remains substantially underdeveloped in current LVLMs. To bridge this gap, we introduce SpatialASM, a scalable framework to

Data video is a media form that integrates data visualization with video narrative, widely adopted in news reporting and business analysis. Compared with general video understanding, data video understanding places greater emphasis on accurately reading data from animated charts, integrating evidence across charts and time, and understanding how narrative organization and visual design communicate information. Yet existing benchmarks target either general videos or static charts, and data video

Knowledge distillation (KD) transfers knowledge from stronger Teacher models to weaker Student models, but most methods require training the Student parameters, thereby binding the distilled knowledge to a specific architecture and checkpoint. This implicit representation is difficult to interpret or reuse across models and limits KD for API-only or costly-to-train models. This paper studies knowledge transfer for large language models (LLMs). We introduce Universal Textual Teaching (UTT), a par

Biological reasoning models use post-training to connect LLMs to biological foundation model representations and biological text. Their benchmark accuracy is taken as evidence that LLMs reason over these inputs. We test this assumption in six biological reasoning models across DNA, protein, and single-cell tasks. We perturb one biological input while holding the query and other inputs fixed, construct evidence conflicts that pair the foundation model representation of one genome, protein, or cel

Meta’s most capable models for personal agents Discussion | Link

Got this couple. It was difficult purchase in this economy but paid only roughly $6100 equivalent for each in my country. I justified the purchase for learning (so I try to convince myself) and coding workflow improvements. I started looking into recipes for GLM 5.3 flash but surprisingly there is very little chatter about that in this sub. The information seem to be scattered and difficult to find. Is the Nvidia forums the best forum to keep updated?