ChatGPT's free tier gets GPT 5.4 mini model with improved coding capabilities
ChatGPT’s free tier gets GPT 5.4 mini model with improved coding capabilities Ben Schoon | Mar 17 2026 - 2:15 pm PT | Mar 17 2026 - 2:15 pm PT OpenAI just…
Search
All sources
xda-developers.com
12
wccftech.com
8
huggingface.co
7
theverge.com
6
9to5mac.com
5
techpowerup.com
5
wired.com
4
9to5google.com
3
techradar.com
3
engadget.com
3
guru3d.com
3
kotaku.com
3
Videos
More videos
[Paper Analysis] On the Theoretical Limitations of Embedding-Based Retrieval (Warning: Rant)
I Spent $600 on Ubiquiti's Industrial Gateway — Here's What You Need to Know
[GRPO Explained] DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
Gl.iNet Mudi 7 WiFI 7 Travel Router - SHOULD YOU BUY?
Energy-Based Transformers are Scalable Learners and Thinkers (Paper Review)
Microsoft Defender can now automatically isolate hacked endpoints
Microsoft Defender can now automatically isolate hacked endpoints By Sergiu Gatlan May 26, 2026 08:19 AM Microsoft is testing a new Defender for Endpoint capability that will automatically isolate compromised endpoints…
E-SMI In-band Library
…Highlights of major release v5.2.1 This release introduces Set and Get operations for enhanced system management capabilities for AMD Family 0x1A and model 0x50-0x5F processors with the following CLI…
I used Claude Code with a local LLM on Ollama, and it’s surprisingly capable for something that's free
…It gets things done, and it’s so capable that you almost start wondering if this thing is alive. But it also burns through credits at an insane rate. You can spend…
Discussions and forums
Show HN: Manage Casio G-Shock watches from the browser
I've been maintaining GShockSmartSync (Android) for a while, and a recurring ask was "can I do this without installing an app?" So I built a browser-based version using the Web Bluetooth API.What it does: - Time sync wit…
2
Ask HN: Are advances in AI going to push Linux to a micro-kernel?
This is something that has been bouncing around my head for the past couple weeks with the flood of security related news around Mythos and the number of 0days being found.Microkernels, unikernals, hardware-enforced capa…
4
9
Ask HN: Is DLSS 4 Multi Frame Generation on RTX 4000 a HW or SW Lock?
NVIDIA markets Multi Frame Generation 3x/4x as an RTX 5000 exclusive feature. But is this actually a hardware limitation, or is it software-enforced?I'm curious if anyone has looked into this. Specifically: - Whether Ad…
1
Google Drive Alternatives with View Only Option for Video Streaming?
I am an adult content creator and I've been hosting my videos on Telegram channels since I started. But it seems that Telegram has been treating me like spam, severely limiting my capability to create channels for my sub…
Is It Just Me, or Has Opus 4.8 Dramatically Improved Session Usage?
I’ve been spending a lot of time coding, debugging, and chatting simultaneously with Opus 4.8, and something interesting stands out: I haven’t hit a session limit yet. With Opus 4.7, my experience was very different. Dur…
Image Generators are Generalist Vision Learners
…capabilities of language understanding and reasoning from generative pretraining. While it has long been conjectured that the ability to create visual content implies an ability to understand it, there has been limited…
Is Switch 2 Powerful Enough To Last Into the 2030s With PS6 and Project Helix Coming Soon?
…Having a good baseline level of hardware and features also mean that, even if image quality or fidelity is limited on a given game versus a higher-end console or PC release…
No, Google's Project Genie Experiment Isn't Trying to be a GenAI Game-Making Kit
…Those tools are not capable of making anything like a Grand Theft Auto VI or Clair Obscur: Expedition 33 . No matter how much they improve, these tools will always be more limited…
Paper page - FocuSFT: Bilevel Optimization for Dilution-Aware Long-Context Fine-Tuning
…This training-time attention dilution (the starvation of content tokens in the attention distribution) weakens the gradient signal, limiting the model's ability to learn robust long-context capabilities. We introduce FocuSFT…
Paper page - BioTool: A Comprehensive Tool-Calling Dataset for Enhancing Biomedical Capabilities of Large Language Models
…A Comprehensive Tool-Calling Dataset for Enhancing Biomedical Capabilities of Large Language Models Published on May 7 Submitted by Xin Gao on May 8 University of California San Diego Authors: , , , , Abstract A…
Paper page - OpenSeeker-v2: Pushing the Limits of Search Agents with Informative and High-Difficulty Trajectories
…Pushing the Limits of Search Agents with Informative and High-Difficulty Trajectories Published on May 5 Submitted by taesiri on May 6 #2 Paper of the day Authors: Yuwen Du , , , , , , Abstract A…