Search: “agents”

Applying Anthropic Primitives at Large Enterprises: Harness Paradigm for Knowledge Work

arXiv:2608.20622v1 Announce Type: new Abstract: Frontier models have collapsed the cost of writing custom code: a niche problem a specialist sees in their own domain now costs an afternoon. The cost o

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Model Releases

[Benchmark] Optimal DFlash2 quants for speed and context size, 5090 RTX, llama.cpp, Qwen 3.8 27B Dynamic3 Unsloth. Comparison with MTP

DGX agent

Graph: A cumulative metric of Speed x Context Size - the higher the dot - the better. Helpful for looking for the most balanced solution. The research relies on DFlash2 Q2 work by AnalogAlok: https://

model-releasesr-localllama

24 Aug 2026

CFM: Language-aligned Concept Foundation Model for Vision

arXiv:2601.13798v3 Announce Type: replace-cross Abstract: Language-aligned vision foundation models perform strongly across diverse downstream tasks. Yet, their learned representations remain opaque,

Dynamic Context Scheduling: Learning Beyond the Static Universe

arXiv:2608.20799v1 Announce Type: new Abstract: We study dynamic context scheduling as a training instrument for contextual re- inforcement learning. Rather than treating intra-episode context variati

Evaluation-as-Search: Adaptive Discovery of Grounding Failures in Meeting Assistants

arXiv:2608.20392v1 Announce Type: cross Abstract: LLM-powered meeting assistants are deployed at scale, yet systematic evaluation of their grounding fidelity remains limited to static benchmarks that

Applications

ExploraTwin, a Non-Profit Research Platform for Digital Twin Simulations

arXiv:2608.20539v1 Announce Type: cross Abstract: Digital twin simulations show promise, but current empirical evidence suggests that the approach should be tested before being deployed in any particu

applicationsarxiv-cs-ai

Fine-Grain GPU Parallelization of the Generalized Partition Crossover for Large-Scale Traveling Salesman Problems

arXiv:2608.21233v1 Announce Type: new Abstract: The Traveling Salesman Problem (TSP) is one of the most extensively studied NP-hard optimization problems. Genetic Algorithm (GA)-based solvers, such as

Infrared Hotspot-Guided Early Warning of Lithium-Ion Battery Thermal Runaway Under Mechanical Abuse

arXiv:2608.20383v1 Announce Type: cross Abstract: Mechanical abuse can trigger thermal runaway (TR) in lithium-ion batteries through localized heat generation before sensor signals become decisive. Th

local-aiarxiv-cs-ai

JuryProbe: An Empirical Consensus-Risk Diagnostic for Routing Reference-Free Factuality Judge Panels to Grounded Verification

arXiv:2608.20607v1 Announce Type: cross Abstract: Panels of inexpensive LLM judges increasingly make accept-or-escalate decisions. In factuality settings, accepting a claim because several reference-f

Just Noticeable Difference Modeling for Token Compression in Vision-Language-Action Models

model-releasesarxiv-cs-cv

arXiv:2608.21247v1 Announce Type: new Abstract: Token compression has become a key technique for reducing the inference cost of large foundation models, with approaches such as token pruning and KV-ca

Lost in Translation: How Universal Ethical Values Fail to Translate Across Global Contexts

arXiv:2608.20490v1 Announce Type: new Abstract: AI ethics frameworks treat values such as fairness, transparency, and accountability as universal and uniformly operationalizable across contexts. We ex

Natural-Language-Guided Generator-Agnostic Shortlisting for Protein Binder Design

arXiv:2608.20755v1 Announce Type: new Abstract: Modern de novo design workflows generate many candidate protein binders, but wet-lab validation capacity remains limited, making shortlisting a major bo

Hardware

Nvidia says its inference accelerator Groq 3 LPX has entered full production and Nebius has signed on as the first customer, and SpaceX will deploy Vera CPUs (Mike Wheatley/SiliconANGLE)

Mike Wheatley / SiliconANGLE: Nvidia says its inference accelerator Groq 3 LPX has entered full production and Nebius has signed on as the first customer, and SpaceX will deploy Vera CPUs — Chipmaker

hardwaretechmeme

PhotoBench: Beyond Visual Matching Towards Personalized Intent-Driven Photo Retrieval

arXiv:2603.01493v2 Announce Type: replace-cross Abstract: Personal photo albums are not merely collections of static images but living, ecological archives defined by temporal continuity, social entan

PromptResponse: Optimizing Prompts for LLM Coding Tasks

arXiv:2608.21074v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used in research workflows and software development pipelines, yet their output remains sensitive to inp

Qwen 3.8 27B, just wanted to say thanks to you guys

I commented on another Qwen 3.8 27B post that I was frustrated getting anything to work. You all gave some great comments. I nuked openwebui and straightened out my llama.cpp docker config. 1 hour of

Six misconceptions about large language models: A minimal model and diagnostic taxonomy

arXiv:2608.20421v1 Announce Type: cross Abstract: Large language models (LLMs) are now embedded in scientific, educational, and governance workflows, with debates centering on their capabilities, mech

SlidesGen-Bench: Evaluating Slides Generation via Computational and Quantitative Metrics

model-releasesarxiv-cs-cl

arXiv:2601.09487v2 Announce Type: replace Abstract: The rapid evolution of Large Language Models (LLMs) has fostered diverse paradigms for automated slide generation, ranging from code-driven layouts

Source-Free MT Evaluation Is Not MT Evaluation

arXiv:2608.20925v1 Announce Type: cross Abstract: Reference-based metrics remain the standard choice in machine translation evaluation, partly because quality estimation methods often correlate less w

StateSight: Benchmarking Latent Spatial-State Reconstruction in Vision-Language Models

arXiv:2608.20414v1 Announce Type: new Abstract: Vision-language models are increasingly used for multimodal question answering, yet their ability to reconstruct latent spatial structure from a single

The Divergence Hypothesis: Unmasking Lexical Interference and Label Bias in Mental Health NLP

arXiv:2608.20353v1 Announce Type: cross Abstract: Computational mental health (CMH) classifiers often degrade under distribution shift because human annotators and distant-supervision pipelines reward

Anthropic’s best AI model struggles to attract users as cheaper tools thrive

model-releasessimon-willison

Anthropic’s best AI model struggles to attract users as cheaper tools thrive A few interesting numbers in this FT story gathered from 'people with knowledge of the matter': Anthropic's 'annualized rev

Best general purpose uncensored or censored coding model with 6GB VRAM and 64GB of RAM?

I haven't had a lot of luck in the reliability of the output of any models so far for local coding use, and find myself relying on Claude's limited free sessions to assist with coding. My coding appli

How would you improve reasoning + memory in a local AI companion?

I'm building a local AI companion and I'm currently working on its cognitive layer. The goal is: User message → understand intent → decide what context is relevant → retrieve only useful memories/stat

local-air-ollama

Quoting Drew Breunig

model-releasessimon-willison

Prior to Fable, it felt silly to waste too much time improving your coding harness or context strategies. A new model would arrive at the same price (or cheaper!) and paper over most of your problems.

Tested in Coding: Q8_K_XL Qwen3.8 27B vs BF16 Qwen3.6 27B

Further to my last post, https://www.reddit.com/r/LocalLLaMA/comments/1vldngi/tested_in_coding_bf16_muse_glimmer_vs_bf16_qwen36/, I bring this requested comparison of Qwen3.8 27B. Context: Both models

“The All Spark” Cluster: Upgrading from 16 - 36 DGX Sparks

Earlier this year I posted about building what at the time I believe was the first 16x DGX Spark Cluster. I’m now adding 20 more Sparks to the cluster in my homelab server rack, giving me 4.6TB of uni

local-air-localllama

I benchmark DFlash 2 (PR build) in llama.cpp on Qwen 3.8 27B against all speculative methods for 3 days. 2.26x on 100 real coding prompts, 4.68x with one n-gram drafter on top. Up to 8x on specific cases.

Hey guys, Inco AI shipped DFlash 2 a few days ago with a drafter for Qwen 3.8 27B and a llama.cpp PR. I built the PR and ran it against plain decoding, MTP, the n-gram lookup drafters, and my July DFl

I forked Ninfer 3090 and converted it to run on the CMP170HX - doubled my Qwen3.6-35B from llama.cpp

Good afternoon, everyone! I wanted to show the work I've been doing around porting Ninfer over to the CMP170HX (Github) So, first, I do want to call out the amazing work that Neroued, Sergiuszm and sp

Ollama + llama.cpp + VS Code + OWUI + ...cloud?

I've been trying to figure out what the best configuration for my local stack would be and have been primarily determining this through trial and error. I wanted to get some outside opinions and stack

model-releasesr-ollama

Industry

Rundoo, a provider of AI-powered business operations software for independent supply stores, raised a $30M Series B led by Battery Ventures (Mike Wheatley/SiliconANGLE)

Mike Wheatley / SiliconANGLE: Rundoo, a provider of AI-powered business operations software for independent supply stores, raised a $30M Series B led by Battery Ventures — Rundoo Inc., the creator of

industrytechmeme

Single RTX 5090: Qwen3.8-27B NVFP4 at a real 262K context in vLLM — 77 tok/s short-context, 64.7 tok/s at 128K

This is the Qwen3.8-27B setup I actually use every day on one RTX 5090. I wanted to write it down with enough detail that another 5090 owner can reproduce it instead of guessing which memory knobs I u

What if Chain-of-Thought wasn’t lossy? Exploring reversible logic (Toffoli/Fredkin-style) for edge LLMs

Right now standard CoT is a one-way street. You generate forward, dump a pile of scratchpad tokens into the KV cache, and pray the model doesn’t hallucinate halfway through. On phones/laptops that cre

local-air-localllama

Bounded Sovereignty and the Control Tax: Pricing AI Oversight When the Deployer Does Not Own the Model

arXiv:2608.19216v1 Announce Type: new Abstract: AI control research asks how to deploy models safely even when they may be misaligned, but many control protocols assume that the deployer can instrumen

Can Conversational AI loosen Us-Versus-Them Boundaries? The Effects of Common, Dual, and Separate Identity Framings on Pro-Immigrant Intergroup Helping

arXiv:2608.19220v1 Announce Type: cross Abstract: Rising immigration has intensified intergroup tensions in many countries. Traditional bias-reduction programs remain difficult to scale and increasing

Electronic Navigational Chart Change Classification

arXiv:2608.20218v1 Announce Type: new Abstract: Electronic Navigational Charts (ENCs) are geospatial vector datasets used in maritime navigation systems that represent hydrographic and navigational in

ExPhy: A Benchmark for Explicit Physical Property Learning in Multi-Object Trajectory Forecasting

arXiv:2608.20009v1 Announce Type: new Abstract: Understanding object dynamics requires not only predicting future trajectories but also examining whether a model captures the physical properties that

FleetSieve: Decision-Critical Profiling for SLO-Aware LLM Fleet Configuration

model-releasesarxiv-cs-lg

arXiv:2608.19659v1 Announce Type: new Abstract: Choosing tensor-parallel (TP) degrees and replica counts for an LLM serving fleet is difficult because performance is not monotonic in TP and the feasib

From Retrieved Context to Runtime Control: Adaptive Compression for Edge-based RAG

arXiv:2608.19535v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) improves language-model responses by grounding generation in external passages, which comes with overhead: retrieve

Growth Without Us: Machine Consumers, Corporate Circularity, and the Decoupling of GDP from Humanity after AGI

arXiv:2608.20231v1 Announce Type: cross Abstract: The standard objection to full automation is demand-side: if humans earn nothing, who buys the output? This confuses an accounting role with a biologi

Inter-X++: A Comprehensive Benchmark for Multimodal Human-Human Interaction Analysis

model-releasesarxiv-cs-cv

arXiv:2608.20312v1 Announce Type: new Abstract: The capability to perceive and synthesize human-human interactions is fundamental to developing intelligent digital human systems. However, existing dat

Is there any interest in a self hosted open source version of manus/perplexity?

I have been working on a project for about 3 months because I found Vane to be awful and I wanted something like Manus for myself. I have a pretty solid app that's probably about ready to release, but

Measuring What a Specification Determines: A Formal Semantic-Block Model and an Execution-Judged Benchmark

arXiv:2608.19475v1 Announce Type: cross Abstract: This work introduces a formal semantic-block model for specifications and an execution-judged benchmark for evaluating specification quality independe

Multimodal API support 🔌 🔹 Set model='deepseek-v4-flash-vision-exp' 🔹 Images are tokenized for billing: up to 384 tokens each, at V4-Flas…

model-releasesdeepseek--x

Multimodal API support 🔌 🔹 Set model='deepseek-v4-flash-vision-exp' 🔹 Images are tokenized for billing: up to 384 tokens each, at V4-Flash pricing 🔹 Supports Chat Completions, Messages & Responses 🔹 S

PL-NBA: A Possession-level Universal Basketball Video Dataset Supporting Multiple Visual Understanding Tasks

model-releasesarxiv-cs-cv

arXiv:2608.19646v1 Announce Type: new Abstract: Visual understanding in sports has emerged as a hot topic in computer vision in recent years. Most existing basketball video datasets adopt single actio

Qwen 3.8 27b is strong even at Q3_xxs

So usually I avoid Q3 quants because I have had bad experiences with it, models were usually too degraded, so the smallest I normally do is Q4, since I only have rtx 4060 ti 16gb. But since there hasn

Qwen3.8-27B at 262K context on a Strix Halo + RTX 3090 Ti: 9.5 -> 153 tok/s, and it beats a dual-3090 vLLM box on HumanEval

Spent a while treating layer placement, KV format and llama.cpp itself as experimental variables. 159 logged experiments. Numbers first, caveats after. Hardware: AMD Ryzen AI MAX+ 395 (Strix Halo, 128

Rethinking the Evaluation and Optimization of LLM-Based Social Simulation

arXiv:2608.19689v1 Announce Type: new Abstract: LLM-based social simulation is a promising complement to traditional methods such as surveys and behavioral experiments. A core question is how to evalu