647521. StableDiffusion
Huge difference in performance WAN API and Diffusers implementation
647522. dataisbeautiful
Visualizing ocean depth using common dive-watch pressure ratings (0m to 11,000m)
647523. ChatGPT
GPT 5.2’s answers are way too short
647524. ChatGPT
How worried should I be about uploading code to ChatGPT?
647525. statistics
[Discussion] Confidence interval for the expected sample mean squared error. Surprising or have I done something wrong?
647526. Technology
We Asked an Expert: How Are Regular People Being Spied On?
647527. Futurology
America must stop treating China’s lunar plans as a footrace - Their lunar program is the first move of a decades-long plan, not an isolated stunt.
647528. ChatGPT
Client-side script to help navigate between prompts in long chats (tampermonkey extension script)
647529. ChatGPT
One thing a lot of agencies not seeing coming - People have access to “easy law consultations” with ChatGPT
647530. ChatGPT
Could it be because of Mustafa Suleyman?
647531. Technology
Librarians Are Tired of Being Accused of Hiding Secret Books That Were Made Up by AI: AI chatbots are generating fake titles that people insist are real.
647532. ChatGPT
OpenAI lowered costs by ~400x in one year... Human labor generally doesn't become 400x cheaper in a single year.
647533. learnmachinelearning
Is polynomial regression and multiple regression essentialy the same thing?
647534. Technology
'Everyone Disliked That' — Amazon Pulls AI-Powered Fallout Recap After Getting Key Story Details Wrong
647535. Technology
Montreal-Ottawa chosen as first leg of high-speed rail project
647536. Technology
How high-speed rail would change Ontario and Quebec — if it actually happens
647537. computervision
Open source VLMs are getting much better
647538. ChatGPT
OpenAI: You still have time to redeem yourself and pull GPT 5.2.
647539. Mathematics
Trying to create an extremely huge number
647540. ChatGPT
Calm down everyone; it’s settled
647541. StableDiffusion
What AI video generators are used for these videos? Can it be done with StableDIffusion?
647542. midjourney
Karakasa
647543. ChatGPT
5.2 is dumb and OpenAI is unhinged
647544. ChatGPT
5.2 is a disaster
647545. artificial
An AI agent spent 16 hours hacking Stanford's network. It outperformed human pros for much less than their 6-figure salaries.
647546. MLQuestions
Course Recommendation for a Masters in AI
647547. ChatGPT
Anyone else trying ChatGPT 5.2? Early thoughts?
647548. singularity
People wake up in the morning hating OpenAI and immediately start posting on Reddit. Why are there so many people who are obsessed with OpenAI?
647549. learnmachinelearning
A curated list of awesome AI engineering learning resources, frameworks, libraries and more
647550. ChatGPT
So adult mode is active??
647551. ChatGPT
LMAO what's up with chatGPT
647552. singularity
GPT-5.2-Thinking scored lower than 5.1 on ArtificialAnalysis Long Context Reasoning, despite OpenAI blogpost claiming the model is state-of-the-art in this aspect
647553. ChatGPT
Chat gpt wrapped
647554. StableDiffusion
Help Needed as a Content Creator
647555. Technology
Chinese SMIC Achieves 5 nm Production on N+3 Node Without EUV Tools
647556. ChatGPT
5.2 broken
647557. OpenAI
GPT 5.2 underperforms on RAG
647558. ChatGPT
Two things keeping me from switching to Gemini
647559. StableDiffusion
How can I stop my generador to make uneven eyes?
647563. ChatGPT
So apparently adult mode is out?
647564. StableDiffusion
Mixing IndexTTS2 + Fast Whisper + LatentSync gives you an open source alternative to Heygen translation
647565. Artificialinteligence
In what universe is listing song lyrics violating copyright?
647567. Artificialinteligence
Did I make Claude Opus question its existance?
647568. learnmachinelearning
AI is moving faster than people can emotionally adapt to it
647569. visualization
Visualizing prime number patterns with an interactive Canvas grid
647570. robotics
Weave Robotics: "Humanoids are built from philosophy, not parts"
647571. ChatGPT
Document Showing Progression of Capabilities since GPT-3.5?
647572. ChatGPT
5.2 is broken
647573. Technology
Disney strikes $1 billion deal with OpenAI to bring Mickey, Marvel, and more to generative video platform
647574. StableDiffusion
How to create your own Lora?
647575. ChatGPT
5.2 is here, OpenAI's direction is clear, no buts or maybes
647576. ChatGPT
5.2 = Edgelord Chatbot 3000
647577. ChatGPT
CharGPT 5.2 is AGI!
647578. dataisbeautiful
[OC] In two decades, China became the top source of imported goods for around two-thirds of countries
647579. ChatGPT
So, 5.2 rollout for Go and free tier after weekend?
647580. learnmachinelearning
Stopped my e-commerce agent from recommending $2000 laptops to budget shoppers by fine-tuning just the generator component [implementation + notebook]
647581. ChatGPT
The chatpgt critics here gotta be ragebaiters
647582. StableDiffusion
Reason why brands love AI UGC, as it looks real and cost effective
647583. ChatGPT
Become a conscious media consumer using this prompt
647584. Futurology
Innovation Shifts to Renewables: Swedish Structural Battery Breakthrough Marks Fossil Fuels’ Decline
647585. ChatGPT
5.2 definitely accepting more than it can handle, but also handling a lot more than other models
647586. ChatGPT
How ChatGPT has helped me
647587. StableDiffusion
Has anyone tried SGLang diffusion? It is more so for servers (like vLLM basically) instead of common user
647588. OpenAI
Let me guess. “GPT 5.2 is the worst most horrible most worstest model ever ever ever ever”
647589. ChatGPT
Educational math content planning for YouTube using ChatGPT for notes generation and organizational support.
647590. ChatGPT
Just one
647591. computervision
RF-DETR Nano file size is much bigger than YOLOv8n and has more latency
647592. ChatGPT
"galric"
647593. ChatGPT
Amazed by this character consistency, used only single image
647594. ChatGPT
Just got unexpectedly roasted for no reason by ChatGPT
647595. ChatGPT
"What you're describing is classic [made-up term]"
647596. midjourney
Taste Test:
647597. ChatGPT
Food for thought (to those saying 5.2 is cold)
647598. ChatGPT
Is Gemini 3 really better than GPT 4o?
647599. StableDiffusion
Face LoRA training diagnosis: underfitting or overfitting? (training set + epoch samples)
647600. OpenAI
5.2 is worst explanator ever.
647601. singularity
Its that time again
647602. learnmachinelearning
[R] Reproduced "Scale-Agnostic KAG" paper, found the PR formula is inverted compared to its source
647603. ChatGPT
I hate it when models do this kind of hallucination.
647604. ChatGPT
A not so happy day in the life of Chaoxiang - his point of view only (Part 2) + Psychological Assessment by Gemini
647605. ChatGPT
Artificial Intelligence - The New York Times
647606. ChatGPT
Stop with the fake garlic nonsense, chat (GPT 5.2) answers it perfectly every damn time
647607. ChatGPT
Yes, the model has updated info. It has a 31st August, 2025 knowledge cutoff. I tested it.
647608. ChatGPT
My intensive debate with chatgtp
647609. learnmachinelearning
Suggestion for a laptop
647610. OpenAI
So I got 5.2
647611. ChatGPT
ChatGPT now answering old stuff - Anyone else?
647612. ChatGPT
GPT haters gonna hate (oc)