All entries

model: support Longcat-Flash (need testing) by ngxson · Pull Request #19182 · ggml-org/llama.cpp

This PR should be ready for testing now. I tested with a very small (8B params) sub-model extracted from the original one. Appreciate if someone can test with the bigger model. GGUF(for testing) from

Model Releases

My first run of Kimi K3 locally.

DGX agent

Running across 2 clusters using llama.cpp over RPC too. Both clusters are not enough to hold everything in memory, so main cluster still partially offloads to run. Goal will be to get all the GPUs in

model-releasesr-localllama

8 Aug 2026

Tools

Neat example here of the agents communicating purely through file names, including adding base64-encoded attachments and using 'zz' prefixes…

Neat example here of the agents communicating purely through file names, including adding base64-encoded attachments and using 'zz' prefixes to ensure their new message sorts to the bottom of the list

toolssimon-willison--x

Now we have a timeline of the OpenAI accidental attack against Hugging Face

model-releasessimon-willison

My comment on Now we have a timeline of the OpenAI accidental attack against Hugging Face — Hacker News.I think one of the most interesting details here might be tucked away in that first bulletin poi

Ollama Cloud reviews

I am wondering if anyone can give opinion on if Ollama Cloud pro or max plans are worth it. Id be looking to use it with Kimi K3, Qwen 3.8 and Deepseek v4flash for now. Wondering if it would be better

model-releasesr-ollama

@OpenAI oo claude code has this now!!! need to try https://x.com/ClaudeDevs/status/2085817074816070014

@OpenAI oo claude code has this now!!! need to try https://x.com/ClaudeDevs/status/2085817074816070014 New in Claude Code: your sessions can now message each other. Instead of having to re-explain you

model-releasesswyx--x

Industry

OpenAI reveals upcoming Astra model may possess ‘critical’ hacking capabilities

OpenAI Group PBC today disclosed that one of its unreleased large language models may pose a significant cybersecurity risk. The algorithm, which is known as Astra, was first detailed last week. OpenA

industrysiliconangle

PSA for anyone with multiple V620's or other gfx1030 cards having problems making llama.cpp tensor split work -- set '-ub 384' and -b to a multiple of that depending on number of GPUs

Basically what the title says. For me, it would always crash and burn trying to use tensor split. Apparently, there's some bug where GPU memory gets corrupted with the default microbatch (512) or high

Local Ai

Quick survey (2 min) on trust in hardware specs for open-source models

Hi everyone, I'm a systems analysis student researching a problem a lot of you probably know well: how much you actually trust the published VRAM/RAM requirements for open-source models before trying

local-air-ollama

Qwen 35B-A3B MoE vs 27B dense in local coding tests: ~4× faster, much smaller quality gap than I expected

I compared Qwen 35B-A3B MoE against Qwen 27B dense on a series of local coding-maintenance tasks. On my R9700/llama.cpp setup, the MoE model generated about 3.9× faster (~116 vs ~30 tok/s), but the co

Qwen3.6 27B + 35B on vLLM, single R9700 (gfx1201)

I've been tuning my new Radeon AI Pro R9700, and figured that this would be useful information for people who are trying to optimise their setups. I'm pretty happy with these results and looking forwa

Showoff Saturday: Local 4x 6000 Pro (multi-year progression)

Not the biggest or shiniest, but it's mine From gaming machine inference on the original llama models, to a 4x RTX 6000 Pro Max Q + 4x 3090s local AI cluster. Pictures are in reverse chronological ord

Local Ai

Spongebob MiniMax H3 test (4 x 5 seconds) turbo 6 steps 1344×768 (AI gen post)

local-air-stablediffusion

MiniMax-H3 in ComfyUI 0.30.0, RTX 4080 16 GB (224 W cap). int8 DiT + int8 Qwen3-VL-32B text encoder. Turbo LoRA @ 0.9, euler + simple, 6 steps, no CFG. MiniMaxH3ReferenceToVideo with 3 reference image

Tesla V100 Qwen3.6 27B Performance

Looking for V100 users to share your config and it's performance. GPU: Tesla V100 PCIE 32Gb Qwen3.6 27B Q4_K_M + Q8_0 MTP 128K context length Pi coding agent llama.cpp model preset: [*] spec-default =

The reports of the demise of Google are greatly exaggerated. I wouldn't underestimate them

model-releasesfrancois-chollet--x

François Chollet commented that claims the demise of Google were greatly exaggerated, cautioning against undervaluation. According to a Polymarket report, Sergey Brin is expected to take direct oversi

We analyzed DeepSeek V4 Flash and GPT-5.6 Luna on DeepSWE. A DeepSeek-first cascade with test-suite verification solved MORE tasks than Luna…

model-releasestogether-ai--x

Researchers from TogetherAI analyzed DeepSeek V4 Flash and GPT‑5.6 Luna on the DeepSWE benchmark. The study found that employing a DeepSeek‑first cascade with test‑suite verification solved more tasks

Weirdly iirc stable diffusion (1.4) finished training around four years ago today too

model-releasesemad-mostaque--x

Emad posted that Stable Diffusion v1.4 reached the end of its training cycle roughly four years before the post was published. Greg Brockman added that GPT‑4 similarly completed training around the sa

When you move a model into production, you want the quality you evaluated to carry through the serving stack. @Kimi_Moonshot benchmarked Kim…

applicationstogether-ai--x

When you move a model into production, you want the quality you evaluated to carry through the serving stack. @Kimi_Moonshot benchmarked Kimi K3 across major inference providers, and Together AI ranke

You may have been told to watch this video about the OpenAI AI hack. You really should, even if you don't usually care about tech stuff. If …

applicationsethan-mollick--x

You may have been told to watch this video about the OpenAI AI hack. You really should, even if you don't usually care about tech stuff. If nothing else, click this link to the 18 minutes in & see how

100% Local RAG Without Internet and Without Ollama

Build a 100% offline fast Retrieval Augmented Generation (RAG) system that runs without an internet connection, without cloud APIs, without OpenAI/Ollama Published a video where you can build a fully

model-releasesr-ollama

~45% lower MiniMax H3 sampler time with new Spectrum settings — degree 1 works surprisingly well (v0.1.8)

model-releasesr-stablediffusion

Follow-up to my original Spectrum MiniMax H3 post: https://www.reddit.com/r/StableDiffusion/comments/1vf1ze3/spectrum_acceleration_for_minimax_h3_in_comfyui/ In that first post, I released the MiniMax

Safety

A Bridge from Audio to Video: Phoneme-Viseme Alignment Allows Every Face to Speak Multiple Languages

arXiv:2510.06612v2 Announce Type: replace Abstract: Speech-driven talking face synthesis (TFS) focuses on generating lifelike facial animations from speech input. Current TFS models perform well in En

safetyarxiv-cs-cv

A Foundational EDM2-Based Generative Model for High-Resolution Synthetic Fetal Ultrasound Imaging from Open Datasets

arXiv:2608.05471v1 Announce Type: cross Abstract: Prenatal ultrasound imaging is key for assessing fetal health, but AI progress is limited by scarce, privacy-restricted, and hard-to-annotate datasets

researcharxiv-cs-cv

A Lexical Analysis of online Reviews on Human-AI Interactions

arXiv:2511.13480v2 Announce Type: replace-cross Abstract: This study focuses on understanding the complex dynamics between humans and AI systems by analyzing user reviews. While previous research has

A llama.cpp PR makes Q2_0 3.0–3.6x faster on x86 CPUs, 8B decode goes 2.39 → 8.20 tok/s

I was going through the current llama.cpp CPU PRs and #26348 stood out because this isn't the usual +5% kernel optimization. It adds an x86 VNNI implementation for the Q2_0 × Q8_0 dot product, and the

Tutorials

A lot of the questions we get from developers are about the concepts behind the API: TTFT, context windows, sampling, fine-tuning, quantizat…

A lot of the questions we get from developers are about the concepts behind the API: TTFT, context windows, sampling, fine-tuning, quantization, deployment tradeoffs. We added Learn to the Together do

tutorialstogether-ai--x

A Low-Power Wearable Respiratory Sensor for Non-Invasive Stress Monitoring

arXiv:2608.05697v1 Announce Type: cross Abstract: Respiration provides a continuously available window into physiological state and behavior. However, monitoring it outside controlled settings remains

researcharxiv-cs-lg

A Master-Salve Robot Manipulator for Needle-Based Teleoperation in MRI Chamber

arXiv:2608.06354v1 Announce Type: new Abstract: We present a MR safe, master-slave robot manipulator for abdominal interventions in the MRI chamber. A human operated 2+1-DoF master controller manipula

researcharxiv-cs-ro

A Multi-Layer System for Ultra-High-Resolution Static 360-Degree Telepresence

arXiv:2608.05570v1 Announce Type: cross Abstract: 360-degree video telepresence offers strong immersive potential but remains constrained by the limited resolution of current capture and display hardw

researcharxiv-cs-cv

A neural operator view on U-Nets for inverse imaging problems

arXiv:2608.05839v1 Announce Type: cross Abstract: Deep neural networks have shown great empirical success in the solution of a wide variety of ill-posed inverse problems in imaging. Yet, very few work

researcharxiv-cs-lg

A note on conditional PAC-efficient reasoning in large language model routing

arXiv:2512.03057v2 Announce Type: replace-cross Abstract: We study distribution-free risk control for model routing, motivated by large language model reasoning. We formalize pointwise conditional eff

A Paragraph is Worth a Thousand Captions: Rethinking Text Supervision for Vision-Language Retrieval

model-releasesarxiv-cs-cv

arXiv:2608.05260v1 Announce Type: new Abstract: Contrastive vision-language models such as CLIP and BLIP are typically trained on short image captions, limiting their ability to retrieve images from d

A Reverse-BSDE Diffusion Sampler

arXiv:2505.06800v2 Announce Type: replace-cross Abstract: Diffusion-based generative models have renewed interest in stochastic differential equation methods for sampling from complex distributions. W

researcharxiv-cs-lg

A Six-Dimensional Taxonomy of Post-Training Adaptation Techniques with Applications in AI Governance

model-releasesarxiv-cs-lg

arXiv:2608.06246v1 Announce Type: new Abstract: Post-training adaptation has become central to modern machine learning practice and includes techniques such as retraining, fine-tuning, parameter-effic

A Study of ASR Adaptation and Representation Dimensionality Reduction in Persian Speech Emotion Recognition Using Whisper

arXiv:2608.05165v1 Announce Type: cross Abstract: Speech Emotion Recognition (SER) in low-resource languages remains a challenging problem due to limited labeled data. In this work, we study the use o

A Survey of Adversarial Efficiency Degradation for Vision Transformer by Exploiting Input-adaptive Optimization

arXiv:2608.05217v1 Announce Type: cross Abstract: Vision Transformers (ViTs) increasingly rely on input-adaptive inference, such as token pruning and early halting, to meet energy and latency budgets.

researcharxiv-cs-cv

Safety

A System for Train Condition Monitoring and Structural Health Assessment of Rail Vehicles

arXiv:2608.05221v1 Announce Type: new Abstract: The ongoing digitalization of rail systems and the increasing use of artificial intelligence (AI) are fundamentally transforming the design, operation,

safetyarxiv-cs-ro

A Two-Tier Perspective on Inference-Time Parallelism in Multi-Agent LLM Systems

model-releasesarxiv-cs-ai

arXiv:2608.05791v1 Announce Type: cross Abstract: Large language model (LLM)-driven multi-agent systems typically require multiple model invocations and complex coordination during inference, and thei

Safety

A Unified Causal Inference Framework for the Desirability of Outcome Ranking Paradigm in Benefit-Risk Evaluation

arXiv:2608.05244v1 Announce Type: cross Abstract: We developed a unified covariate-adjusted causal inference framework for estimating the desirability of outcome ranking (DOOR) probability for benefit

safetyarxiv-cs-lg

A Unified Framework for Trajectory Prediction with Explicit Planning and Reaction Decomposition

arXiv:2608.05673v1 Announce Type: new Abstract: Trajectory prediction has shifted toward structured formulations with explicit social modeling. However, existing methods inadequately distinguish the f

A Unified Risk View of Uncertainty: Posterior Risk for Disentanglement and Evaluation Beyond Proxies

model-releasesarxiv-cs-lg

arXiv:2608.05995v1 Announce Type: new Abstract: Reliable uncertainty estimates are critical in safety-sensitive applications, where understanding the sources of predictive uncertainty is essential. Th

Local Ai

A visualization of LLM API costs to ask for local resources

I have not been successful with management to get funding for local resources despite bringing forth solid arguments about data sovereignty and related architectures. What actually succeeded in gettin

local-air-localllama

ABC: Numerical Data Collection under Local Differential Privacy without Prior Knowledge

arXiv:2608.05737v1 Announce Type: cross Abstract: Local Differential Privacy (LDP) provides strong privacy guarantees for collecting numerical data. A fundamental challenge, however, is that existing

Agents

absolute @FireworksAI_HQ domination for anybody wondering what happens to the inference platforms when hyperscalers compete... here's a nice…

absolute @FireworksAI_HQ domination for anybody wondering what happens to the inference platforms when hyperscalers compete... here's a nice narrative violation to chew on :) 📣 @Kimi_Moonshot's Kimi K

agentssonya-huang--x

Abstract Event Causal Rules: Induction and Application

model-releasesarxiv-cs-ai

arXiv:2608.05205v1 Announce Type: new Abstract: Event-centric intelligent analytical systems heavily depend on explicit causal event knowledge for risk early warning, decision-making support and narra

Accelerating nanodrug development in continuous flow systems using informed prediction models based on low-cost surrogate nanoparticles

arXiv:2608.05761v1 Announce Type: new Abstract: The development of nanotherapeutics often involves extensive empirical optimization due to the sensitivity of nanoparticle properties, such as size and

applicationsarxiv-cs-lg

Accurate Localization of Road Traffic Objects on the Road Plane Using Surveillance Camera Imagery

arXiv:2608.05840v1 Announce Type: new Abstract: Accurate vehicle localization from monocular roadside surveillance cameras is important for intelligent transportation systems, traffic monitoring, and

applicationsarxiv-cs-cv

Agents

Acoustic-driven millimetric helical robot: ultrasonic synergistic manipulation in confined fluidic environment

arXiv:2608.05746v1 Announce Type: new Abstract: Acoustic field-driven manipulation provides a non-contact and non-invasive strategy for controlling microscale and nanoscale objects, yet its extension

agentsarxiv-cs-ro