model releases

Qwen 3.8 27B running on a single RTX 5090 researches and creates a full animation using only code.

The goal of this post is 2 fold. First, I wanted to show what a local model like Qwen 3.8 27B running on a single RTX 5090 GPU in a harness like Row-Bot can do. Second, I wanted to create a full featu

model-releasesr-ollama

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Model Releases

Qwen-3.8-Flash-Next on 1x RTX 5090: TG=50 t/s, PP=2300 t/s - with FreeToken

DGX agent

I was somewhat disappointed with the performance of Qwen 3.8 Flash Next on my single RTX 5090 using llama.cpp. One issue is that llama.cpp still has no Expert Caching implemented for MoE models. There

model-releasesr-localllama

19 Sept 2026

Qwen3.8-27B at 144 tok/s on an M5 Max MacBook Pro

Meet Inco Splash, open-source inference engine, built around the model and around Apple silicon. Up to 3× the decode speed of Ollama, 2× oMLX, and almost 4× when an agent fans out into sub-agents. Req

really great thread/benchmark for common sense in robotics and a reminder that we are not actually close to AGI

model-releasesgary-marcus--x

really great thread/benchmark for common sense in robotics and a reminder that we are not actually close to AGI GPT-6 Astra attempted harmful actions 97% of the time when it was asked to stab a human-

Recently, Datalab released OmniExtractBench, a benchmark that combines several benchmarks (our own ExtractBench, LongExtractBench, other ven…

model-releasesjerry-liu--x

Recently, Datalab released OmniExtractBench, a benchmark that combines several benchmarks (our own ExtractBench, LongExtractBench, other vendors) to measure structured document quality. The benchmark

Reproduce it, or it doesn't count: why training-side decontamination can't be verified, and what an evaluation-side rule looks like [D]

model-releasesr-machinelearning

Since OpenAI retired SWE-bench Verified in February (every frontier model tested could reproduce reference fixes for some tasks; underspecified tests rewarded knowing the intended fix), I've been tryi

Steer LLMs and Agents at the Token Level: An interactive tool for token visualization & control, model inspection and data annotation.

onPanda is designed for geeks, power users, curious minds, and engineers. Its UI is built for deep exploration and efficient data annotation. - The core loop is simple: hover over a token → click an a

Training a Neural Network on AMD MI50s Using Vulkan: Proof of ConceptOr: Why I Stopped Listening and Just Did It

Note: This writeup was put together with the help of AI. So I've got dual AMD MI50 32GB cards. If you know these cards, you know the story - AMD dropped official ROCm support for gfx906 after ROCm 5.7

Tuning Qwen 3.8 27B and OMP as a coding agent on 2× 3090s

Oh My Pi + vLLM on two 3090s. Average wait per turn went from 28s to 7s, mostly from changing omp settings: explicit effort level on every role (unset ones defaulted to xhigh) thinking_token_budget of

v0.34.3

model-releasesollama-releases

What's Changed GET /api/show now advertises each model's thinking controls and default: curl http://localhost:11434/api/show -d '{'model': 'glm-5.3-flash:cloud'}' { 'thinking': { 'values': ['low', 'hi

Weekend project, getting AEO-pilled. Also a good test (for me) of Claude Projects. Have been meaning to update Eleventy and do a few other r…

Weekend project, getting AEO-pilled. Also a good test (for me) of Claude Projects. Have been meaning to update Eleventy and do a few other redesign tasks on https://cybercultural.com, which tbh has be

model-releasesswyx--x

6 billion tokens for $65 Is it as good as Astra and Fable at hard stuff? No of course not. But you don't always need a genius and can just c…

model-releasesemad-mostaque--x

6 billion tokens for $65 Is it as good as Astra and Fable at hard stuff? No of course not. But you don't always need a genius and can just call one for advice when you do. Extrapolate where this is go

768gb vram for less than the price of one RTX 6000

I have always posted about budget builds on here, and often asked how we are going to run the next big models. Often Plenty of downvotes too or folks telling me that it's not running if I'm getting 5t

A Network Science Approach to Granular Time Series Segmentation

model-releasesarxiv-cs-lg

arXiv:2505.17640v3 Announce Type: replace Abstract: Time series segmentation assigns a label to each part of a sequence. We formulate dense univariate segmentation as node classification on a graph wh

A Neuropsychologically Grounded Evaluation of LLM Cognitive Abilities

arXiv:2603.02540v2 Announce Type: replace Abstract: Large language models (LLMs) display a unified 'general factor' of capability across 10 benchmarks (a finding confirmed by our factor analysis of 15

A Smaller Transformer in Your Transformer

model-releasesarxiv-cs-cv

arXiv:2609.20100v1 Announce Type: new Abstract: Recent findings indicate that Vision Transformers settle into locally similar computational phases, implying a level of depthwise computational redundan

A Unified Evaluation Framework for Trustworthy Large Language Models, Agentic AI, and Multimodal Systems

arXiv:2609.19524v1 Announce Type: new Abstract: Benchmark scores alone provide an incomplete basis for assessing the trustworthiness of modern artificial intelligence systems. Large language models (L

Accelerating the borderless Lakehouse: Announcing preview of cross-cloud caching

model-releasesgoogle-cloud-ai

Today, we are excited to announce enhancements to the borderless Lakehouse, our answer to how data engineers, data scientists, and increasingly, AI agents, can query governed data directly where it li

Affective Shared Autonomy: Temporal Affect Dynamics and Subjective Evaluation in Bimanual Teleoperation Tasks

model-releasesarxiv-cs-ro

arXiv:2609.19802v1 Announce Type: new Abstract: Physical teleoperation integrates human cognitive flexibility with robotic precision, yet demanding manipulation tasks frequently induce severe cognitiv

Agentic AI Networking for Heterogeneous Unmanned Aerial Systems in Low-Altitude Wireless Networks

arXiv:2609.19538v1 Announce Type: new Abstract: Low-altitude wireless networks (LAWNs) are emerging as a key infrastructure for heterogeneous unmanned aerial systems that support concurrent services w

AgriScope: Pixel-Grounded Multimodal Understanding for Agricultural Images

model-releasesarxiv-cs-cv

arXiv:2609.20325v1 Announce Type: new Abstract: Agricultural image understanding requires fine-grained recognition of plant diseases, pests, crop structures, and botanical species under complex real-w

ALIBI: Adversarial Legitimacy Injection in Binary Input against LLM Malware Analyzers

model-releasesarxiv-cs-lg

arXiv:2609.19722v1 Announce Type: cross Abstract: Large language models are being integrated into malware triage workflows as reasoning components that summarize static evidence and produce analyst-fa

Amazon SageMaker Inference: 2026 year-to-date launches in review

model-releasesaws-ml-blog

Amazon SageMaker AI shipped 13 inference launches in the first half of 2026 across two deployment paths: fully managed endpoints and Amazon SageMaker HyperPod Inference. This post reviews each launch,

An Analysis of Training-Free Self-Reported Confidence in Language Models

model-releasesarxiv-cs-cl

arXiv:2609.20541v1 Announce Type: new Abstract: Large language models can report a numerical confidence together with generated content, but it is unclear whether this report is more than calibrated r

An Architecture for Long-Horizon Agents: Levels, Ticks and Cascaded Intelligence

arXiv:2609.19519v1 Announce Type: new Abstract: Language-model agents are increasingly asked to carry out work spanning days or weeks, such as an operations remediation or a research programme. Such a

An Empirical Study of Harness Design for Coding Agents

arXiv:2609.20804v1 Announce Type: new Abstract: Coding harnesses shape how autonomous coding agents translate model capabilities into long-horizon software-engineering performance, yet existing work t

Anthropic adds support for the AGENTS.md instructions spec to Claude Code; OpenAI contributed AGENTS.md to the Agentic AI Foundation last year (Thomas Claburn/The Register)

Thomas Claburn / The Register: Anthropic adds support for the AGENTS.md instructions spec to Claude Code; OpenAI contributed AGENTS.md to the Agentic AI Foundation last year — Devs cheer relief from c

model-releasestechmeme

Anthropic opens AI-powered biology research lab

model-releasessiliconangle

Anthropic PBC has opened a wet lab, a facility dedicated to biology research, in the San Francisco Bay Area. Reuters reported today that the company will use robots to automate certain scientific task

AnyviewMeter: Adapting Robotic Reward Models with Camera Geometry and Multi-View Attention

model-releasesarxiv-cs-cv

arXiv:2609.20106v1 Announce Type: new Abstract: Robotic reward models evaluate task execution from visual observations, but their predictions can change with camera viewpoint and occlusion even when t

Architectural Design, Not Only Model Intelligence, Governs Multi-Agent LLM Performance

arXiv:2602.03128v2 Announce Type: replace Abstract: Multi-agent LLM frameworks are data-intensive systems that govern how agents orchestrate tasks, manage state, and coordinate decisions. These archit

AUDITPLAN: Commit, Then Answer for Auditable Safety Alignment

arXiv:2609.19325v1 Announce Type: cross Abstract: Safety tuning pipelines judge only the final answer, which makes it difficult to distinguish robust refusal from two undesirable shortcuts: blanket re

Automated Gradient-Driven Parameter Sharing for Low-Resource Multilingual Speech-to-Text Translation

model-releasesarxiv-cs-cl

arXiv:2603.25836v2 Announce Type: replace Abstract: In low-resource multilingual speech-to-text translation, uniform architectural sharing across languages frequently introduces representation conflic

AVTrace: Diagnosing Audio-Visual Temporal Reasoning in Omni Models

arXiv:2609.19991v1 Announce Type: cross Abstract: Omni models can describe video content, but can they locate events in time, preserve event order, and judge audio-visual synchronization? We introduce

b11029

vulkan: raise the hoisted row-id limit for mul_mat_id from 256 to 1024 experts (#28501) vulkan: raise the hoisted row-id limit for mul_mat_id to 512 experts The expert-count shader (count_experts.comp

b11030

ci : bump android-actions/setup-android to 4.0.4 (#29065) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/48395907 macOS/iOS: macOS Apple Silicon (arm64) ma

b11033

cmake : fix build when GGML_CPU=OFF and GGML_CUDA=ON (#29026) fix: build fails when GGML_CPU=OFF and GGML_CUDA=ON fix: eol in examples/convert-llama2c-to-ggml/CMakeLists.txt file Website: https://llam

b11034

vocab : add ufakzeka pre-tokenizer (#29033) vocab : add ufakzeka pre-tokenizer vocab : move ufakzeka to the models list and regenerate the hash mapping Website: https://llama.app Attestations: https:/

b11035

vulkan: add IQ3_S MMQ matmul kernels (#28822) vulkan: add IQ3_S MMQ matmul kernels Make block_a_to_shmem do 2-byte loads (110 bytes is divisible by 2) Align the check, IQ3_S is also using K tile size

b11036

ggml : handle graph buffer reservation failure (#26070) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/48431150 macOS/iOS: macOS Apple Silicon (arm64) macO

b11037

ggml-webgpu: fix supports_op condition for GET_ROWS (#28978) fix get_rows vec4 handling Add src strides checking to vec4_aligned of get_rows and the new test case. Website: https://llama.app Attestati

b11039

Model-Saver: Write the SWA pattern, 15 more architectures roundtrip (#29042) llama: read the SWA pattern as a period or a per-layer array Add llama_model_base::load_swa_pattern(), which reads sliding_

b11040

ggml : check for allocation failures to prevent crashes (#28149) ggml : check for allocation failures to prevent crashes wording Website: https://llama.app Attestations: https://github.com/ggml-org/ll

b11042

opencl: add bin kernel kernel_gemm_noshuffle_q6_k_f32_32b_trans_ila_a8_bin (#28678) opencl: add A8 Q6_K non-MoE binary kernel opencl: fix layout compatibility Website: https://llama.app Attestations:

b11043

hexagon: HMX flash-attention head_dim padding (support DK=DV=72) (#26539) Allow HMX flash-attention to run with head_dim not a multiple of 64 (e.g. SigLIP head_dim=72), by operating on DK/DV rounded u

b11044

hexagon: im2col update (#29103) ggml-hexagon: accept 1D and padded IM2COL ops ggml-hexagon: make pure-DDR IM2COL kernel is_2D-aware ggml-hexagon: extend IM2COL DMA patch-embed fast path to 1D ggml-hex

b11045

hexagon: add ROLL op support (#29105) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/48565469 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple Silicon (a

b11046

opencl: add support for bin kernel flash_attn_f32_f16_bin (#29046) opencl: add flash_attn_f32_f16_bin opencl: guarded prefill fa Website: https://llama.app Attestations: https://github.com/ggml-org/ll

Before the Warning Comes Too Late: Incremental Phone-Scam Detection from Speech

model-releasesarxiv-cs-cl

arXiv:2609.20223v1 Announce Type: new Abstract: We study weakly supervised incremental telecom fraud detection from raw telephone audio, where training provides only conversation-level labels and pred