model releases

Engineer / developer observations of Gemma4-31B, Qwen3.8-27B, and 6.1-Sol for software engineering work

For what it's worth. I'm a mechanical engineer by education and software engineer in practice over the past ~20 years. Playing around with local models has been a recent hobby. Today I decided to do a

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Model Releases

Finetuned Qwen3.5-4B into a decision model like Jev using @UnslothAI. - 4-bit QLoRA, bf16, single A100, 15 min. - Train/val loss flatten aft…

DGX agent

Finetuned Qwen3.5-4B into a decision model like Jev using @UnslothAI. - 4-bit QLoRA, bf16, single A100, 15 min. - Train/val loss flatten after 2 epochs so most of the learning happens fast. - Trained

model-releasesclem-delangue--x

10 Oct 2026

Fully local conversational AI: Whisper + Hermes 8B + Kokoro, zero cloud, running inside a plush toy

A conversational AI plush toy where nothing leaves my network. Everything runs on my LAN: STT: Whisper LLM: Hermes 3 Llama 3.1 8B (Q8_0) TTS: Kokoro Memory: per-person (LangGraph). It knows who it's t

honestly pretty cool to have this on my new tab page just merged in some PRs by others too!

honestly pretty cool to have this on my new tab page just merged in some PRs by others too! you now get Claude API credits with your MAX plans ($100, or 200 matching your plan) every month, use this t

model-releasesthariq--x

How to make gemma 4 works with SWEBench?

I wanted to compare gemma 4 31b qat q4 and qwen3.8 27b iq4_nl, so I run SWEBench verfied on them. The original gemma 4 chat template doesn't work with preserve reasoning, so google released an update

I asked 3 Qwen 3.8 models to solve recent AtCoder problems at 4 thinking efforts

I was looking into finalizing my daily driver and it was not clear which one to chose. I have a 5060ti 16GB so I picked the following contenders, which seem to be working at a reasonable speed on my h

I have an ESC4000 G3 with 8x T4s in it - what is the fastest way I can deploy Qwen3.5-9B for about 10-15 users concurrently: currently using llama.cpp

Hi All: I have an Asus ESC 4000 G3 with 128 GB DDR4 RAM - I tried putting in my V620s but couldn’t put more than 2. Sadly, pivoted to T4s, these are 72 watt passive cards and I thought I could use the

I have no insight into the facts of this situation, but “CLAIM: (outline a claim)/FACT: (outline facts that don’t address the claim just lai…

model-releasesanthropic--x

**Summary** * Daniel Dale (@ddale8) comments that the common structure of a claim/fact‑check—“CLAIM: (outline a claim)/FACT: (present unrelated facts)”—is not how reliable fact checks operate. * He e

I tried three ways of getting Gemma and Qwen to update a notes file in Ollama

I ran a small test for Dory to see whether a local model could update one fact in a synthetic notes file without losing any of the others. I asked Gemma and Qwen to rewrite the file, return a patch, o

model-releasesr-ollama

Improve token per second without touching quant

Spent the past month tweaking and experimenting with many different numbers to achieve 30tps. Hardware: -Rx6700xt 12gb vram (AMD) -2x16 ddr4 3200 ram -r5 5600x -llama.cpp vulkan sdk -window11 (no wsl

Introducing Anatomy Describe a machine in one line. Claude brings it to life. - Interactive isometric wireframes you can rotate, inspect and…

model-releasesyohei-nakajima--x

Introducing Anatomy Describe a machine in one line. Claude brings it to life. - Interactive isometric wireframes you can rotate, inspect and explore - WebGL shaders for fire, water, plasma and more -

Mistral OCR is on OpenDocRouter! For its price, it is decent over tables and reading order. It's also blazing fast at p50 1.8s per page We'r…

model-releasesjerry-liu--x

Mistral OCR is on OpenDocRouter! For its price, it is decent over tables and reading order. It's also blazing fast at p50 1.8s per page We're constantly bringing in new models since OpenDocRouter laun

Mreaia v1.0.0: Universal Zero-Token Engine & Hardware-Locked Context Distillation for LLMs (>99.9% Reduction)

Hey everyone! 👋 Feeding multi-gigabyte files (SQL dumps, videos, audio, PDFs, and spreadsheets) directly into LLMs often causes context window overflow, GPU VRAM bottlenecks, massive token bills, and

model-releasesr-ollama

Open-source Mac app that runs EmbeddingGemma 2 locally to search your files by what’s in them

DigUp is a free Mac app that runs Google DeepMind’s new EmbeddingGemma 2 locally over your own files. The model puts text, images, audio and video in one space, so you describe what you remember and l

Qwen 3.6 35B A3B: 131K context + vision on 6GB VRAM

Running HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive Q4_K_M with llama.cpp at ~600 tok/s prefill and 23 tok/s decode, 131k context window, Q8 KV cache - on an RTX 2060 6GB + 32GB DDR4 RAM.

Qwen Flash Next @ 137 tok/s & 3,497 tok/s Prefill w/ 512k context on a 5090, 192gb ram, Windows Build, comparing Strata and Infernix

Strata vs Infernix on the same model, RTX 5090 — A/B at 262k and 512k context Hi folks, another inference engine post for your feed. I was blown away by Strata, but saw someone else post about Inferni

Qwen3.8-27B on a single 3090: 140 tok/s on code with a custom megakernel

I've been using Claude Opus 5.5 to speed up Qwen3.8-27B on my PC (rtx 3090), it wrote a CUDA megakernel that is 1.4-1.9x faster than llama.cpp depending on the task/context length. The results and cod

Sources: Nvidia is in talks to acquire or invest more into US open-weights AI startup Reflection AI; the deal may be an acquihire to avoid antitrust scrutiny (Financial Times)

Financial Times: Sources: Nvidia is in talks to acquire or invest more into US open-weights AI startup Reflection AI; the deal may be an acquihire to avoid antitrust scrutiny — Donald Trump's administ

model-releasestechmeme

Strata with Qwen3.8 Flash Next UD-Q4_K_XL

Been seeing quite a few posts about Strata lately, so I figured I'd give it a shot on my RTX PRO 6000 and I am very impressed. Most of the benchmarks I've seen are using IQ2 or IQ3 quants, so I wanted

The ultimate LLM benchmark: Chilean Spanish. 80% of AI models fail to recognize Chilean Spanish and insist on speaking like an Argentine

model-releasescristobal-valenzuela--x

The ultimate LLM benchmark: Chilean Spanish. 80% of AI models fail to recognize Chilean Spanish and insist on speaking like an Argentine 80% de modelos de IA no logra reconocer el español chileno e in

We @Sequoia are thrilled to back @TypesafeAI @CompleteSkeptic. ChatGPT put AI front and center in the consumer mind. Claude Code transformed…

model-releasessonya-huang--x

We @Sequoia are thrilled to back @TypesafeAI @CompleteSkeptic. ChatGPT put AI front and center in the consumer mind. Claude Code transformed the day-to-day work of the developer/knowledge worker. Jev

3BASiL: An Algorithmic Framework for Sparse plus Low-Rank Compression of LLMs

arXiv:2603.01376v2 Announce Type: replace Abstract: Sparse plus Low-Rank (mathbf{S} + mathbf{LR}) decomposition of Large Language Models (LLMs) has emerged as a promising direction in model compressio

4-Tensor Attention Model for Semantic Physical Reality

model-releasesarxiv-cs-cl

arXiv:2610.11716v1 Announce Type: cross Abstract: We describe a 4-tensor attention model that predicts the next semantic state of a scene, for video generation and robot planning. A window of states h

A 3D Characterization Framework for Intelligent Sequential Decision Making

arXiv:2610.11696v1 Announce Type: new Abstract: Puzzles are widely used to evaluate the reasoning capabilities of artificial intelligence (AI) systems for sequential decision making, yet approaches or

A Camera-Native Stereo VR180 Dataset

model-releasesarxiv-cs-cv

arXiv:2610.10607v1 Announce Type: new Abstract: Immersive VR180 video is increasingly produced with professional stereo fisheye cameras, yet public VR180 research resources are mostly collected from o

A Closer Look at Agentic BBO: Benchmarking LLM Agents for Black-Box Optimization

arXiv:2610.12183v1 Announce Type: cross Abstract: Black-box optimization (BBO) arises in many scientific and engineering problems where objective evaluations are expensive and limited. Recent large la

A Dual-Hypothesis Reasoning Framework for LLM Guardrails

arXiv:2607.17575v3 Announce Type: replace Abstract: We propose ARBITER, a novel LLM guardrail framework that introduces two key ideas: (i) dual-hypothesis reasoning, a reasoning method for LLM guardra

A mesh-based neural energy method for the simulation of heterogeneous composites

arXiv:2610.10862v1 Announce Type: cross Abstract: Modeling heterogeneous materials remains a challenge for physics-informed neural networks such as the deep energy method (DEM). The DEM and its varian

A new feature for my blog, built using my voice

model-releasessimon-willison

I shipped a new feature for my blog today: the Newsletters page, which offers an index of all of the newsletters I've sent out, both my free weekly Substack and my monthly sponsors-only updates. I bui

A Survey on LLM-Integrated Hardware Design Verification

arXiv:2610.10580v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly being integrated into hardware verification to automate specification interpretation, verification-artif

A Unified Score Matching Paradigm for Video Anomaly Detection and Anticipation

model-releasesarxiv-cs-cv

arXiv:2610.11149v1 Announce Type: new Abstract: Video anomaly detection (VAD) is a fundamental and safety-critical task in computer vision. Recent generative approaches detect anomalies from a distrib

A week of free access to Qwen3.8-Max, Qwen3.8-Flash, and Wan3.0 on @gmi_cloudis live. Explore what you can build with us now. 🙌

A week of free access to Qwen3.8-Max, Qwen3.8-Flash, and Wan3.0 on @gmi_cloudis live. Explore what you can build with us now. 🙌 We're extending FREE Qwen 3.8 Max + Flash for 7 more days! And we've inc

model-releasesqwen--x

Accurate and Efficient Object Pose Estimation via the Aggregation of Diffusion Features

model-releasesarxiv-cs-cv

arXiv:2403.18791v4 Announce Type: replace Abstract: Estimating the pose of objects from images is a crucial task of 3D scene understanding, and recent approaches have shown promising results on very l

ActionCodec: What Makes for Good Action Tokenizers

arXiv:2602.15397v2 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) models leveraging the native autoregressive paradigm of Vision-Language Models (VLMs) have demonstrated superior

AdaCast: Conditional Parameter Generation for Adaptive Time Series Forecasting

arXiv:2610.12240v1 Announce Type: cross Abstract: Time-series foundation models (TSFMs) have achieved strong forecasting performance across domains. However, most adaptation methods remain static. Exi

Adapting English Quality Classifiers for Multilingual LLM Pretraining Data Selection

arXiv:2610.11585v1 Announce Type: cross Abstract: Recent advances in large language model (LLM) pretraining highlight the role of high-quality training data in improving performance. While model-based

AdaptLSTM: Efficient Adaptive Online Learning for Cloud Workload Forecasting under Distribution Drift

arXiv:2610.12265v1 Announce Type: new Abstract: Accurate workload forecasting is critical for elastic resource provisioning in web-scale cloud services, where distribution shifts driven by viral conte

Agent4RE: A Self-Refining Multi-agent Framework for End-to-End Software Requirements Engineering and Benchmarking

arXiv:2610.10628v1 Announce Type: cross Abstract: Existing LLM-based approaches for software Requirements Engineering (RE) typically rely on basic prompting strategies or rudimentary agent collaborati

AgentHorizon: Evaluating Agentic Judges for Long-Horizon Computer-Use Tasks

arXiv:2610.11050v1 Announce Type: new Abstract: Computer-use agents are capable of completing complex tasks, increasing the use of automatic judges to determine success, either for training or for eva

Agentic-TTT: Training test-time policy for test-time training

arXiv:2610.12002v1 Announce Type: cross Abstract: Test-time training (TTT) adapts an LLM's parameters using signals derived from test inputs, and can make striking improvements in pre-specified settin

AgentTime: Can Agents Estimate and Control Their Own Runtime?

arXiv:2610.09944v2 Announce Type: replace Abstract: An essential control of AI agents is their ability to manage runtime. This ability requires a sense of time-awareness, to predict and estimate wall-

Amortized Off-Policy Evaluation for LLMs

arXiv:2610.10848v1 Announce Type: new Abstract: Accurate evaluation is central to selecting which LLM to deploy, yet testing a candidate on live traffic exposes real users to an unvetted model. Teams

An Explainable Header-Centric Framework for Large-Scale Semantic Table Interpretation and Data Quality Assessment

arXiv:2610.10541v1 Announce Type: new Abstract: Knowledge Graph (KG) quality depends not only on downstream graph validation, but also on the quality of tabular metadata used before integration. In me

Anthropic’s AI gave Philadelphia police a fake tip about an unsolved homicide

model-releasesthe-verge-ai

An Anthropic AI model provided false information about an unsolved homicide to a Philadelphia Police Department (PPD) tipline, according to a report from 6abc. In a statement released on Friday, the P

Arbiter: Detecting Interference in LLM Agent System Prompts

arXiv:2603.08993v2 Announce Type: replace-cross Abstract: System prompts for LLM-based coding agents are software artifacts that govern agent behavior, yet lack the testing infrastructure applied to c

ASPIRE: Saddle-Point Discovery through Set Prediction and Physical Refinement

arXiv:2610.10969v1 Announce Type: new Abstract: Predicting thermally activated diffusion and defect evolution with event-driven models requires identifying atomic rearrangement mechanisms and their ac

b11516

model-releasesllama-cpp-releases

hexagon: fix IM2COL patch-embed DMA ring overflow (#30189) hexagon: fix IM2COL patch-embed DMA ring overflow The exact-tiling (stride == kernel, no pad/dilation) IM2COL DMA kernel issues ICKH DDR->VTC

b11517

model-releasesllama-cpp-releases

CUDA: pass src1 precision to host MMQ config helpers (#30168) CUDA: pass src1 precision to host MMQ config helpers Signed-off-by: ynankani ynankani@nvidia.com make src1 prec explicit Signed-off-by: yn