GPT 5 The Verdict From Power Users And Critics
A Mixed Reception The Community Reacts to GPT 5
The launch of GPT-5 has been met with a whirlwind of discussion, particularly from the power users and developers who rely on these models daily. While some see it as an incremental step forward, others feel it's a significant downgrade, sparking debates about performance, cost, and OpenAI's overall strategy. A common sentiment is that while GPT-5 might be faster, the quality of its output, especially for complex reasoning, has declined compared to previous models like o3.
One user notes, "I swear I had an understanding of how to get deep analytical thinking out of o3. I am absolutely struggling to get the same results with GPT-5." Others report technical issues, such as the model abruptly losing the thread of a conversation, feeling "like talking to someone who wasn't listening."
Is GPT 5 a Cost Cutting Strategy
A prominent theory circulating within the community is that the new GPT-5 is fundamentally a cost-cutting measure for OpenAI. With a goal of scaling to a billion users on a GPU-intensive product, the argument is that OpenAI is using a routing layer to direct queries to less powerful, cheaper models to manage costs. One commentator states, "I still think GPT5 is really a cost cutting measure... The real story is running these models at true performance max likely could go into the thousands per month per user. And so we are being constrained."
This sentiment is echoed by users who feel the performance of the paid $20/month plan has been deliberately throttled. Another user points out, "Has anyone noticed that OpenAI is cutting off submitted context in ChatGPT Pro? ... This seems quite unethical given they advertise 128k context, and I doubt it's an accident (since it runs in the direction of cost savings)."
The Secret Power of GPT 5 Pro
Complicating the narrative is the mention of a separate, more powerful model: GPT-5 Pro. Several users who claim to have tested it report that it is a true state-of-the-art model, far outperforming competitors like Grok 4 Heavy and Opus 4.1. One user who spends six hours a day with these models shared a specific success story: "I have a bug that was a complex interaction between backend and front end over websockets. The day before I was banging my head against the wall with o3 pro and grok heavy, gpt5 solved it first try."
However, access to this Pro model seems limited, and its performance isn't universally agreed upon. Another power user offered a more tempered view: "My initial impressions are that 5-pro is maybe 0-2% more knowledgeable and 5-10% more inventive/original than o3-pro." This suggests that even the top-tier models may be offering only marginal gains, or that their capabilities are highly dependent on the specific task.
Performance Showdown GPT 5 vs The Competition
When it comes to specific tasks, the reviews are varied. For analytical tasks, formal logic, and data analysis, GPT-5 Pro is described as "superhuman and state-of-the-art." However, for creative writing and prose, some users find it to be a "poor model, obviously and transparently worse than Kimi K2 and Deepseek R1." One user makes a striking claim: "It never ceases to amaze me that the best English prose stylists are the Chinese models... Kimi is actually on par with most published poets."
Claude 3 Opus is also praised for its stylized prose, though it has its own quirks. This highlights a growing trend where users must switch between different models to find the best tool for a specific job, rather than relying on a single, all-powerful AI.
The Coding Conundrum Mixed Reviews from Developers
For developers, the experience is a mixed bag. Some, like the user who fixed a complex websocket bug, find GPT-5 Pro to be a lifesaver. Others feel it's a step back. One developer lamented, "my personal feeling gpt5-thinking is much faster but doesnt produce the same quality results as o3 which were capable to scan through the code base dump with file names and make correct calls."
A recurring and frustrating issue is hallucination. A user trying to mod Cyberpunk 2077 found that "ChatGPT 5 just hallucinated some APIs even after doing 'research' and repeatedly being called out." This fundamental problem of reliability remains a major hurdle. The single biggest improvement, many agree, would be for the model to simply say, "I don't know."
The Great AI Debate Hype vs Reality
The GPT-5 launch has also fueled a broader discussion about the state of AI. Many users are tired of the hype. One comment reads, "Lying and hype have replaced genuine innovation. It’s sad that lying and pushing nonsense is ‘part of the game’ because it shouldn’t be." Critics like Gary Marcus are seen by some as essential voices for integrity and transparency, while others dismiss them as contrarians.
Ultimately, the launch reflects a maturing but fragmented market. As one user aptly put it, "People had grown to expect miracles, but GPT-5 is just the latest incremental advance." The focus for OpenAI may be shifting from pure model performance to building an integrated product ecosystem to maintain its lead in consumer adoption, even if its technical edge is no longer as clear.
To Pay or Not to Pay The Subscription Dilemma
The rapid evolution of AI models has led some to question the value of subscriptions. "It's absolutely a bad idea to buy a subscription for any of the models right now," one user argued, noting that offerings become outdated within months. Others find the $20/month fee a small price to pay for the productivity boost and the convenience of not losing context by switching between free models. The business model itself is under scrutiny, with some predicting that AI providers will have to move away from simple subscriptions and potentially introduce advertising or other monetization strategies to survive the "race to the bottom."
