Browse
60 stories from 90 sources
1) #2 Gemini-2.5-Pro-Preview-05-06 (Score: 2) #3 GLM-4.5 (Score: 2)…Chatbot Arena Elo Rankings — Top 20 Models
1 Gemini-2.5-Pro (Score: 1) #2 Gemini-2.5-Pro-Preview-05-06 (Score: 2) #3 GLM-4.5…
1 Gemini-2.5-Pro (Score: 1) #2 Gemini-2.5-Pro-Preview-05-06 (Score: 2) #3 GLM-4.5…
AI is only part of the storyThe Trends That Will Actually Matter in the Next Decade
AI is only part of the story.
AI is only part of the story.
someone asked me to post this here, said you gays would like…Bare-Metal AI: Booting Directly Into LLM Inference ‚ No OS, No Kernel (Dell E6510)
someone asked me to post this here, said you gays would like this kinda thing.
someone asked me to post this here, said you gays would like this kinda thing.
Previously This benchmark continues my local testing on personal…Qwen3 Coder Next | Qwen3.5 27B | Devstral Small 2 | Rust & Next.js Benchmark
Previously This benchmark continues my local testing on personal production repos,…
Previously This benchmark continues my local testing on personal production repos,…
My frends trained and benchmarked 4 diffusion model versions…My frends trained and benchmarked 4 diffusion model versions entirely on an RTX 2050 (4GB VRAM) — the 17.8M model beat the 143.8M one
Reddit LocalLLaMA: My frends trained and benchmarked 4 diffusion model versions…
Reddit LocalLLaMA: My frends trained and benchmarked 4 diffusion model versions…
I've asked gemini 3.1 what it thinks about humanity chance that we…Ai alignment - the great filter
I've asked gemini 3.1 what it thinks about humanity chance that we can solve the…
I've asked gemini 3.1 what it thinks about humanity chance that we can solve the…
Polymarket has been allowing people to bet on when the US would…Polymarket defends its decision to allow betting on war as ‘invaluable’
Polymarket has been allowing people to bet on when the US would strike Iran next.
Polymarket has been allowing people to bet on when the US would strike Iran next.
OpenAI: Our agreement with the Department of WarOpenAI: Our agreement with the Department of War
Reddit singularity: OpenAI: Our agreement with the Department of War
Reddit singularity: OpenAI: Our agreement with the Department of War
POV: Claude watching ChatGPT take hours to eliminate Ayatollah…POV: Claude watching ChatGPT take hours to eliminate Ayatollah Khamenei. meanwhile Claude smoked Maduro in just 15 minutes
Reddit singularity: POV: Claude watching ChatGPT take hours to eliminate Ayatollah…
Reddit singularity: POV: Claude watching ChatGPT take hours to eliminate Ayatollah…
I thought I'd share my experience with Qwen3.5Qwen3.5 family running notes
I thought I'd share my experience with Qwen3.5.
I thought I'd share my experience with Qwen3.5.
Acing this new AI exam — which its creators say is the…Acing this new AI exam — which its creators say is the toughest in the world — might point to the first signs of AGI
Reddit Artificial: Acing this new AI exam — which its creators say is the toughest in…
Reddit Artificial: Acing this new AI exam — which its creators say is the toughest in…
Industry expectations in Machine Learning Engineers in 2026Industry expectations in Machine Learning Engineers in 2026
Reddit MachineLearning: Industry expectations in Machine Learning Engineers in 2026
Find the core claim, method, and released artifacts.
how do you guys handle all these AI subscriptionsHow do you handle all these AI subscribtions?
how do you guys handle all these AI subscriptions?
how do you guys handle all these AI subscriptions?
Support for dict attribute is a little inconsistent in Dynamo: Support dict in NestedUserFunctionVariable ()
Support for dict attribute is a little inconsistent in Dynamo.
Support for dict attribute is a little inconsistent in Dynamo.
Hey everyone, I have been building this on this side for a…I built a tool to automate your workflow after recording yourself doing the task once (Open Source)
Hey everyone, I have been building this on this side for a couple of months now and…
Hey everyone, I have been building this on this side for a couple of months now and…
"Cancel ChatGPT" movement goes mainstream after OpenAI closes deal…"Cancel ChatGPT" movement goes mainstream after OpenAI closes deal with U.S. Department of War — as Anthropic refuses to surveil American citizens
Reddit singularity: "Cancel ChatGPT" movement goes mainstream after OpenAI closes deal…
Reddit singularity: "Cancel ChatGPT" movement goes mainstream after OpenAI closes deal…
Anthropic CEO Dario Amodei sat down with CBS News for an exclusive…Exclusive interview: Anthropic CEO Dario Amodei on Pentagon feud
Anthropic CEO Dario Amodei sat down with CBS News for an exclusive interview, hours…
Anthropic CEO Dario Amodei sat down with CBS News for an exclusive interview, hours…
Feb 28 , 18:34 UTC Resolved - Between 9:50 PT / 17:50…Elevated errors on Claude Opus 4.6
Feb 28 , 18:34 UTC Resolved - Between 9:50 PT / 17:50 UTC and 10:12 PT / 18:12 UTC we…
Feb 28 , 18:34 UTC Resolved - Between 9:50 PT / 17:50 UTC and 10:12 PT / 18:12 UTC we…
Claude #1 in CanadaClaude #1 in Canada
Reddit singularity: Claude #1 in Canada
Open the primary source for full details.
I feel like that could be the sweet spot for 64GB VRAM…Is anyone else waiting for a 60-70B MoE with 8-10B activated params?
I feel like that could be the sweet spot for 64GB VRAM, and could reach the performance…
I feel like that could be the sweet spot for 64GB VRAM, and could reach the performance…
I feel like everything in the AI industry is spedrunning profit…This sub is incredible
I feel like everything in the AI industry is spedrunning profit driven vendor lock in…
I feel like everything in the AI industry is spedrunning profit driven vendor lock in…
In light of recent events, I recommend canceling your Chatgpt…Cancel your Chatgpt subscriptions and pick up a Claude subscription.
In light of recent events, I recommend canceling your Chatgpt subscription and picking…
In light of recent events, I recommend canceling your Chatgpt subscription and picking…
There's been a lot of buzz about Qwen3.5 models being smarter than…Qwen3.5 35B-A3B replaced my 2-model agentic setup on M1 64GB
There's been a lot of buzz about Qwen3.5 models being smarter than all previous…
There's been a lot of buzz about Qwen3.5 models being smarter than all previous…
https://reddit.com/link/1rh95cl/video/0b8dqf83x9mg1/player # The…How I built multi-agent AI system where agents peer-review each other before I approve
https://reddit.com/link/1rh95cl/video/0b8dqf83x9mg1/player # The setup that shouldn't…
https://reddit.com/link/1rh95cl/video/0b8dqf83x9mg1/player # The setup that shouldn't…
Even with newer models like GPT-5.2 and Claude 4.6, AI chatbots still…Even frontier LLMs from GPT-5 onward lose up to 33% accuracy when you chat too long
Even with newer models like GPT-5.2 and Claude 4.6, AI chatbots still give worse…
Even with newer models like GPT-5.2 and Claude 4.6, AI chatbots still give worse…
Really interesting projectTiny transformers (<100 params) can add two 10-digit numbers to 100% accuracy
Really interesting project.
Really interesting project.
Hi All, I just want to share that I distilled the LAION…AudioMuse-AI-DCLAP - LAION CLAP distilled for text to music
Hi All, I just want to share that I distilled the LAION CLAP model specialized for…
Hi All, I just want to share that I distilled the LAION CLAP model specialized for…
If you've used multi-agent setups with LangChain, CrewAI, AutoGen,…What if LLM agents passed KV-cache to each other instead of text? I tried it -- 73-78% token savings across Qwen, Llama, and DeepSeek
If you've used multi-agent setups with LangChain, CrewAI, AutoGen, or Swarm, you've…
If you've used multi-agent setups with LangChain, CrewAI, AutoGen, or Swarm, you've…
AGI fanboys, do you still want agiWe are officially entering in beta version of Skynet. AI can't replicate itself, yet.
AGI fanboys, do you still want agi?
Open the primary source for full details.
Hey all — quick question for senior PhD folksAI/ML PhD Committee
Hey all — quick question for senior PhD folks.
Hey all — quick question for senior PhD folks.
new google paper is out and it challenges something a lot of…google found that longer chain of thought actually correlates NEGATIVELY with accuracy. -0.54 correlation
new google paper is out and it challenges something a lot of us assumed.
new google paper is out and it challenges something a lot of us assumed.
Welcome to the weekend, friendsYou can still grab great deals on Bose headphones and Astro Bot this weekend
Welcome to the weekend, friends!
Welcome to the weekend, friends!
The first technique that pushed gpt-oss-20b to 3 refusals from 100…Multi-Directional Refusal Suppression with Self-Organizing Maps - Pull Request into heretic!
TL;DR: The first technique that pushed gpt-oss-20b to 3 refusals from 100 while keeping…
TL;DR: The first technique that pushed gpt-oss-20b to 3 refusals from 100 while keeping…
Feb 28 , 15:50 UTC Resolved - This incident has been resolvedElevated errors on claude.ai
Feb 28 , 15:50 UTC Resolved - This incident has been resolved.
Feb 28 , 15:50 UTC Resolved - This incident has been resolved.
qwen3.5 35b-a3b evaded the zero-reasoning budget by doing its…qwen3.5 35b-a3b evaded the zero-reasoning budget by doing its thinking in the comments
Reddit LocalLLaMA: qwen3.5 35b-a3b evaded the zero-reasoning budget by doing its…
Reddit LocalLLaMA: qwen3.5 35b-a3b evaded the zero-reasoning budget by doing its…
A Dream of Spring for Open-Weight LLMs: 10 Architectures from…A Dream of Spring for Open-Weight LLMs: 10 Architectures from Jan-Feb 2026
Reddit MachineLearning: A Dream of Spring for Open-Weight LLMs: 10 Architectures from…
Find the core claim, method, and released artifacts.
How reusable, lazy-loaded instructions solve the context bloat…Claude Skills and Subagents: Escaping the Prompt Engineering Hamster Wheel
How reusable, lazy-loaded instructions solve the context bloat problem in AI-assisted…
How reusable, lazy-loaded instructions solve the context bloat problem in AI-assisted…
Still I see some folks mentioning models like Qwen-2.5, Gemma-2,…Why some still playing with old models? Nostalgia or obsession or what?
Still I see some folks mentioning models like Qwen-2.5, Gemma-2, etc., in their threads…
Still I see some folks mentioning models like Qwen-2.5, Gemma-2, etc., in their threads…
I know everyone has their own subjective take on what models are…Qwen 3.5-35B-A3B is beyond expectations. It's replaced GPT-OSS-120B as my daily driver and it's 1/3 the size.
I know everyone has their own subjective take on what models are the best, at which…
I know everyone has their own subjective take on what models are the best, at which…
Xiaomi has announced its first Bluetooth tracker, and while the…Xiaomi’s tracker doesn’t need a case to clip to your keys
Xiaomi has announced its first Bluetooth tracker, and while the Xiaomi Tag has a more…
Xiaomi has announced its first Bluetooth tracker, and while the Xiaomi Tag has a more…
Xiaomi and Leica's long-running phone partnership just got a little…Xiaomi’s Leica Leitzphone mostly earns the name
Xiaomi and Leica's long-running phone partnership just got a little closer.
Xiaomi and Leica's long-running phone partnership just got a little closer.
Xiaomi has just given a global launch to two of its latest…Xiaomi 17 is a small(ish) phone with a big(ish) battery
Xiaomi has just given a global launch to two of its latest flagship phones, the Xiaomi…
Xiaomi has just given a global launch to two of its latest flagship phones, the Xiaomi…
What are your thoughts on the OpenAI deal with DOWWhat are your thoughts on the OpenAI deal with DOW?
Reddit singularity: What are your thoughts on the OpenAI deal with DOW?
Reddit singularity: What are your thoughts on the OpenAI deal with DOW?
In preparation for an XPU-specific backend for scaledmmv2 , move…: Factor out scaled_mm algo checks to non-CUDA ()
Summary: In preparation for an XPU-specific backend for scaledmmv2 , move some helpful…
Summary: In preparation for an XPU-specific backend for scaledmmv2 , move some helpful…
Flow matching is often discussed in the context of image generation…Works on flow matching where source distribution comes from dataset instead of Gaussian noise?
Flow matching is often discussed in the context of image generation from Gaussian noise.
Flow matching is often discussed in the context of image generation from Gaussian noise.
We introduce MediX-R1, an open-ended Reinforcement Learning (RL)…MediX-R1: Open Ended Medical Reinforcement Learning
We introduce MediX-R1, an open-ended Reinforcement Learning (RL) framework for medical…
We introduce MediX-R1, an open-ended Reinforcement Learning (RL) framework for medical…
We present a scalable 3D reconstruction model that addresses a…VGG-T: Offline Feed-Forward 3D Reconstruction at Scale
We present a scalable 3D reconstruction model that addresses a critical limitation in…
We present a scalable 3D reconstruction model that addresses a critical limitation in…
Numerous lines of aim to control $\textit{model disagreement}$ --…Model Agreement via Anchoring
Numerous lines of aim to control $\textit{model disagreement}$ -- the extent to which…
Numerous lines of aim to control $\textit{model disagreement}$ -- the extent to which…
We identify occlusion reasoning as a fundamental yet overlooked…SeeThrough3D: Occlusion Aware 3D Control in Text-to-Image Generation
We identify occlusion reasoning as a fundamental yet overlooked aspect for 3D…
We identify occlusion reasoning as a fundamental yet overlooked aspect for 3D…
A dataset server must often distribute the same large payload to many…A Dataset is Worth 1 MB
A dataset server must often distribute the same large payload to many clients,…
A dataset server must often distribute the same large payload to many clients,…
Bio-inspired event cameras have recently attracted significant…Sensor Generalization for Adaptive Sensing in Event-based Object Detection via Joint Distribution Training
Bio-inspired event cameras have recently attracted significant research due to their…
Bio-inspired event cameras have recently attracted significant research due to their…
The Platonic Representation Hypothesis posits that neural networks…SOTAlign: Semi-Supervised Alignment of Unimodal Vision and Language Models via Optimal Transport
The Platonic Representation Hypothesis posits that neural networks trained on different…
The Platonic Representation Hypothesis posits that neural networks trained on different…
The lack of reasoning capabilities in Vision-Language Models (VLMs)…Scale Can't Overcome Pragmatics: The Impact of Reporting Bias on Vision-Language Reasoning
The lack of reasoning capabilities in Vision-Language Models (VLMs) has remained at the…
The lack of reasoning capabilities in Vision-Language Models (VLMs) has remained at the…
Standard mixed-precision training of neural networks requires many…FlashOptim: Optimizers for Memory Efficient Training
Standard mixed-precision training of neural networks requires many bytes of accelerator…
Standard mixed-precision training of neural networks requires many bytes of accelerator…
Coarse data arise when learners observe only partial information…Mean Estimation from Coarse Data: Characterizations and Efficient Algorithms
Coarse data arise when learners observe only partial information about samples; namely,…
Coarse data arise when learners observe only partial information about samples; namely,…
Open-vocabulary segmentation (OVS) extends the zero-shot…Retrieve and Segment: Are a Few Examples Enough to Bridge the Supervision Gap in Open-Vocabulary Segmentation?
Open-vocabulary segmentation (OVS) extends the zero-shot recognition capabilities of…
Open-vocabulary segmentation (OVS) extends the zero-shot recognition capabilities of…
Recent advances in machine learning have emphasized the integration…Differentiable Zero-One Loss via Hypersimplex Projections
Recent advances in machine learning have emphasized the integration of structured…
Recent advances in machine learning have emphasized the integration of structured…
AI-powered scientific research tools are rapidly being integrated…Understanding Usage and Engagement in AI-Powered Scientific Research Tools: The Asta Interaction Dataset
AI-powered scientific research tools are rapidly being integrated into research…
AI-powered scientific research tools are rapidly being integrated into research…
Neural network accelerators have been widely applied to edge…Bitwise Systolic Array Architecture for Runtime-Reconfigurable Multi-precision Quantized Multiplication on Hardware Accelerators
Neural network accelerators have been widely applied to edge devices for complex tasks…
Neural network accelerators have been widely applied to edge devices for complex tasks…
A growing number of publications address the best practices to use…Utilizing LLMs for Industrial Process Automation
A growing number of publications address the best practices to use Large Language…
A growing number of publications address the best practices to use Large Language…
No stories found
Try adjusting your search or filters