Comment by dsco
17 hours ago
DeepSeek Flash really came at a bad time before the Anthropic IPO. I use this thing daily and it’s neck and neck with SOTA models. The search is one of the things setting it apart.
17 hours ago
DeepSeek Flash really came at a bad time before the Anthropic IPO. I use this thing daily and it’s neck and neck with SOTA models. The search is one of the things setting it apart.
It's absolutely not SOTA for vibe coding a whole app, but for guided work it is great.
Is anyone actually oneshotting a whole app or even vibecoding one in multiple steps without it being "guided work"? I really wish announcements (like DeepSeek's recent one) would stop focusing on the ability of the model to spit out Minecraft clones on demand.
Some people are trying to one shot entire companies
7 replies →
OP didn’t mention one-shotting.
There are plenty of use cases for the ability to quickly through together and refine an application without reviewing the code though.
I have one-shotted the two most recent relatively complex apps I've tried:
1. a puzzle game with complex interplay, pieces to place, design vs. play mode, storing multiple solutions with the puzzle, and more. I've since iterated something like 200 times as I worked out the design etc., but the first iteration was 99% correct and what I asked for
2. a learn-to-read application with built-in stories (that the LLM wrote) the ability to import epub, text, and HTML, that uses the built-in speech libraries as a default, but also supports an open-source TTS library that I gave in the first prompt, tracks progress and problem workds, has multiple games to reinforce learning, and has a parental mode for tracking progress and configuring the tool overall.
Frankly I was stunned that Claude one-shotted that second app, there was a lot in the prompt, and it did it all. I fully expected to have to iterate on the first draft multiple times. But it worked out of the box.
With Opus 5.5 you can vibe code a whole app without reading the code at all. Obviously you still have to tell it what to make.
With earlier models that doesn't really work. You really have to read the code and watch it like a hawk or it will make a big mess.
To put it another way, with lesser models you have to guide them at a code level. With Opus 5.5 you only have to guide it at an architectural / feature level.
3 replies →
which model is SOTA for vibe coding a whole app? I always use models+agents in guided setup
the moment there is a model which can code whole app, we can fairly say software engineering is done.
Opus 5.5 + Claude Code (or probably any agent harness; I don't think they make that much difference). I wouldn't say software engineering is done quite yet but yeah it's definitely never going to be the same.
It's really not neck and neck with Opus 5.5.
Cost-adjusted,it might be ahead.
It's key to name the boundary conditions. Unlimited cost is a very HN assumption.
Max Subscription seems to be the HN assumption most of the time
Really depends on your use case.
You are right that there's plenty of use cases where Opus 5.5 is better.
If it depends on your use case then it’s not neck and neck…
1 reply →
True, it isn't neck-and-neck. It's neck and balls. Deepseek's balls and Opus 5.5's neck. Whatever you can build with Opus 5.5, I can build in the same time, 1/500th the price with Deepseek Flash 4.1.
Or at a really good time, depending on who you ask.
Not everyone wants Anthropic to succeed.
If Anthropic or OpenAI fails, the other wins the US market, no? I’m oversimplifying but…that’s a pretty realistic possibility in the face of people thinking sinking one US AI company would hurt the greater US venture into AI.
> If Anthropic or OpenAI fails, the other wins the US market, no?
I am of the opinion that they both either succeed or they both fail. They are both losing money hand over fist and their model quality is pretty close to each other. There is not a whole lot that sets them apart.
2 replies →
What if both fail?
6 replies →
No? Why would either of one of those have to win? Even if you limit yourself to US companies only, Google and Meta still exist.
The others, plural, meaning Gemini, Muse, and all the open weight models hosted by a competitive landscape of US providers. There is plenty of American players in the space outside of OpenAI and Anthropic if that’s the only thing you care about.
If Anthropic and OpenAI economics are as bad as they seem to be they deserve to fail, others will take over the remaining market. It’s not a team sport
i thought open models weren't SOTA. therefore not neck and neck... in your experience, do you think it depends on your type of work?
They're about half a gen behind. Very impressive to get this kind of utility from a flash series model... the impact of RL seems to be behind much of the coding and agentic improvements lately.
In my experience, for coding, they are pretty much SOTA. Maybe not for other sciences and problem solving.
Models are comming every 4-6 months. Anthropic, OpenAI, all of them know this.
4-6 months? Try between 8-18 days for a foundation lab release.
It literally is not though... even DeepSeek doesn't claim that. It's just super cost-effective
> bad time
Regardless of the performance metrics... why bad time?
Deepseek is more like a pet project with huge state backing, and they don't seem to compete with Anthropic in terms of profitability.
Deepseek is profitable.
Anthropic is most likely not. Despite bilions of state funding and contracts.