Ask HN: Are OpenAI models IP-range restricted?

2 lynx97 0 7/22/2025, 5:32:57 AM

I oserve that the availability of OpenAI models seems to be tool and/or IP-range restricted. Does anyone else observe something similar?

I have NOT verified my "organisation". So I can not use "llm -m o3", OK. However, at the university, I can use o3 and o3-pro with aider. No errors. Same API key. At home, I can not use o3-pro with aider, error is roughly "chat/completion does not support this model, did you use the wrong endpoint?". However, aider with o3 works at home. "llm -m o3" does not work at home. I sacually checked the aider source code for some magic which would make o3 available while it does not work with "llm -m o3". Didn't find anything. This is rather strange. Does anyone else experience similar things?

Seven Effective TikTok Content Strategies for Entrepreneurs

Show HN: InkyCut – The open-source Canva alternative with a vibe editor (inkycut.com)

NASA hacked hardware of camera orbiting Jupiter (theregister.com)

Trump admin wants to curtail 'gain-of-function' research (theconversation.com)

Identity Crisis in the Aftermath of Human-Decentering (doppioperfavore.substack.com)

DeepMind Table Tennis Robots Train Each Other (spectrum.ieee.org)

The Hater's Guide to the AI Bubble (wheresyoured.at)

Bell Labs Takes a Topological Approach to Quantum 2.0 (nextplatform.com)

Anatomy of a SYN-ACK Attack (akamai.com)

Show HN: AI Powered Mock Interview Practice Platform (app.vivaquest.in)

Notable People Globe (tjukanovt.github.io)

Show HN: Arc OS – spec-first audit layer for GPT prompts (MD and self-check) (github.com)

We Won't Reach AGI by Scaling Up LLMs (youtube.com)

Comparing the Glove80 and Maltron Keyboards (tratt.net)

How to Firefox (kau.sh)

The modern art form that redeemed a Windows utility has lessons for all (theregister.com)

Support Amazon Multi-Site Operations for Cross-Border E-Commerce

X refuses to cooperate with French probe into algorithm bias (lemonde.fr)

Status, Class, and the Crisis of Expertise (conspicuouscognition.com)

The Em Dash Responds to the AI Allegations (mcsweeneys.net)

Ways to Donate to the FSF (fsf.org)

DreamHAT+ Enables 60 GHz Radar Sensing on Raspberry Pi 4B and 5 (linuxgizmos.com)

Jazz Herald (jazzherald.de)

Paying to pollute: How carbon offsets raised emissions in China (voxdev.org)

We maintain HarfBuzz – Ask us anything (or tell us what confused you) (old.reddit.com)

Information is Beautiful Awards 2024 (informationisbeautifulawards.com)

A Man Who Killed Google Search (wheresyoured.at)

Show HN: Simple Background Job with PostgreSQL (dataqueue.dev)

A 'data sufficiency' mindset is key for responsible research (nature.com)

Show HN: Desktop YouTube Music with Native UI and Plugin SDK (github.com)

Practical tips on building LLM agents (letters.lossfunk.com)

A Self-Hosted, BSD-Native Gemini Protocol Server Stack (journal.bsd.cafe)

Mixture of Recursions vs. Transformers: Efficiency Unlocked (algogist.com)

$11,000 Chinese EV with solid state batteries is coming soon (electrek.co)

We Built a Language Model 14,000,000x Smaller Than GPT3 and Formally Verified It (github.com)

Show HN: KeyVocab – drill vocab while you type (github.com)

Subsetter – Web Development-focussed CLI tool for generating font subsets (github.com)

Foreword for Fuzz Testing Book (pages.cs.wisc.edu)

Patrick Soon-Shiong Taking LA Times Public [video] (youtube.com)

LLM Just Got Pwned: The Hidden Danger in MCP Sampling (yassworks.github.io)

Nvidia Canary-Qwen-2.5B (huggingface.co)

Charities/clubs: Submit your loyalty card (scan-my-card.com)

The imminent death of HTTP/1.1 and its risks are not understood (mikhailian.mova.org)

Claude-Flow v2.0.0 Alpha (github.com)

Using Vision Language Models to Index and Search Fonts (lui.ie)

Possible Origins of Peculiar Canaanite Rituals (archaeology.org)

NAPI-RS 3.0 Released (napi.rs)

GNU Taler (en.wikipedia.org)

Weak password allowed hackers to sink a 158-year-old company (bbc.com)

When a meteor smashes into your driveway (popsci.com)

Ask HN: Are OpenAI models IP-range restricted?

Comments (0)