I Tested DeepSeek’s R1 and V3 Coding Skills – and we’re not All Doomed (Yet).

DeepSeek exploded into the world’s awareness this past weekend. It stands out for 3 powerful reasons:

1. It’s an AI chatbot from China, rather than the US

2. It’s open source.

3. It uses significantly less infrastructure than the big AI tools we’ve been looking at.

Also: Apple scientists expose the secret sauce behind DeepSeek AI

Given the US government’s concerns over TikTok and possible Chinese federal government participation because code, a new AI emerging from China is bound to produce attention. ZDNET’s Radhika Rajkumar did a deep dive into those issues in her article Why China’s DeepSeek might break our AI bubble.

In this article, we’re avoiding politics. Instead, I’m putting both DeepSeek V3 and DeekSeek R1 through the same set of AI coding tests I have actually tossed at 10 other big language designs. According to DeepSeek itself:

Choose V3 for tasks needing depth and precision (e.g., fixing sophisticated math problems, producing complicated code).

Choose R1 for latency-sensitive, high-volume applications (e.g., customer support automation, fundamental text processing).

You can select in between R1 and V3 by clicking the little button in the chat interface. If the button is blue, you’re utilizing R1.

The short response is this: remarkable, however clearly not ideal. Let’s dig in.

Test 1: Writing a WordPress plugin

This test was actually my very first test of ChatGPT’s programming prowess, method back in the day. My wife needed a plugin for WordPress that would help her run a participation device for her online group.

Also: The very best AI for coding in 2025 (and what not to use)

Her needs were relatively simple. It needed to take in a list of names, one name per line. It then had to arrange the names, and if there were replicate names, separate them so they weren’t listed side-by-side.

I didn’t truly have time to code it for her, so I chose to give the AI the challenge on an impulse. To my big surprise, it worked.

Ever since, it’s been my first test for AIs when examining their shows skills. It needs the AI to understand how to set up code for the WordPress structure and follow prompts clearly sufficient to produce both the interface and program logic.

Only about half of the AIs I’ve tested can completely pass this test. Now, nevertheless, we can include another to the winner’s circle.

DeepSeek V3 produced both the interface and program logic exactly as specified. When It Comes To DeepSeek R1, well that’s a fascinating case. The “reasoning” element of R1 triggered the AI to spit out 4502 words of analysis before sharing the code.

The UI looked different, with much larger input locations. However, both the UI and logic worked, so R1 also passes this test.

Up until now, DeepSeek V3 and R1 both passed among four tests.

Test 2: Rewriting a string function

A user complained that he was unable to get in dollars and cents into a contribution entry field. As written, my code just allowed dollars. So, the test includes providing the AI the regular that I wrote and asking it to reword it to permit both and cents

Also: My preferred ChatGPT function just got way more effective

Usually, this results in the AI creating some regular expression validation code. DeepSeek did generate code that works, although there is room for improvement. The code that DeepSeek V2 wrote was unnecessarily long and repetitious while the thinking before producing the code in R1 was likewise really long.

My greatest issue is that both designs of the DeepSeek validation ensures validation approximately 2 decimal locations, but if a huge number is entered (like 0.30000000000000004), using parseFloat does not have specific rounding knowledge. The R1 model likewise utilized JavaScript’s Number conversion without inspecting for edge case inputs. If bad information returns from an earlier part of the regular expression or a non-string makes it into that conversion, the code would crash.

It’s odd, due to the fact that R1 did provide a really good list of tests to verify against:

So here, we have a split decision. I’m offering the point to DeepSeek V3 since neither of these concerns its code produced would cause the program to break when run by a user and would create the anticipated results. On the other hand, I need to provide a fail to R1 since if something that’s not a string somehow enters into the Number function, a crash will occur.

And that gives DeepSeek V3 2 triumphes of 4, however DeepSeek R1 just one triumph of 4 up until now.

Test 3: Finding a bothersome bug

This is a test produced when I had a very annoying bug that I had problem locating. Once again, I chose to see if ChatGPT could manage it, which it did.

The obstacle is that the response isn’t apparent. Actually, the difficulty is that there is an apparent answer, based on the mistake message. But the obvious response is the incorrect response. This not only captured me, but it routinely catches a few of the AIs.

Also: Are ChatGPT Plus or Pro worth it? Here’s how they compare to the free version

Solving this bug needs comprehending how particular API calls within WordPress work, having the ability to see beyond the error message to the code itself, and after that knowing where to find the bug.

Both DeepSeek V3 and R1 passed this one with nearly identical answers, bringing us to 3 out of 4 wins for V3 and two out of 4 wins for R1. That already puts DeepSeek ahead of Gemini, Copilot, Claude, and Meta.

Will DeepSeek score a home run for V3? Let’s discover out.

Test 4: Writing a script

And another one bites the dust. This is a tough test since it requires the AI to understand the interplay between three environments: AppleScript, the Chrome object model, and a Mac scripting tool called Keyboard Maestro.

I would have called this an unfair test because Keyboard Maestro is not a traditional programming tool. But ChatGPT dealt with the test easily, understanding precisely what part of the issue is managed by each tool.

Also: How ChatGPT scanned 170k lines of code in seconds, conserving me hours of work

Unfortunately, neither DeepSeek V3 or R1 had this level of knowledge. Neither model knew that it required to split the job between directions to Keyboard Maestro and Chrome. It likewise had fairly weak understanding of AppleScript, composing custom regimens for AppleScript that are native to the language.

Weirdly, the R1 design stopped working as well since it made a bunch of inaccurate presumptions. It presumed that a front window always exists, which is certainly not the case. It likewise made the assumption that the presently front running program would always be Chrome, rather than clearly checking to see if Chrome was running.

This leaves DeepSeek V3 with three proper tests and one fail and DeepSeek R1 with 2 right tests and 2 fails.

Final thoughts

I found that DeepSeek’s insistence on utilizing a public cloud email address like gmail.com (rather than my regular e-mail address with my corporate domain) was frustrating. It likewise had a variety of responsiveness stops working that made doing these tests take longer than I would have liked.

Also: How to utilize ChatGPT to write code: What it does well and what it doesn’t

I wasn’t sure I ‘d be able to compose this post due to the fact that, for many of the day, I got this error when trying to register:

DeepSeek’s online services have recently faced large-scale harmful attacks. To ensure ongoing service, registration is momentarily limited to +86 contact number. Existing users can visit as normal. Thanks for your understanding and support.

Then, I got in and had the ability to run the tests.

DeepSeek seems to be excessively loquacious in regards to the code it generates. The AppleScript code in Test 4 was both incorrect and excessively long. The routine expression code in Test 2 was right in V3, but it could have been composed in a manner in which made it a lot more maintainable. It failed in R1.

Also: If ChatGPT produces AI-generated code for your app, who does it truly come from?

I’m certainly satisfied that DeepSeek V3 beat out Gemini, Copilot, and Meta. But it seems at the old GPT-3.5 level, which suggests there’s certainly room for enhancement. I was disappointed with the outcomes for the R1 model. Given the option, I ‘d still select ChatGPT as my shows code helper.

That stated, for a new tool running on much lower infrastructure than the other tools, this could be an AI to watch.

What do you think? Have you tried DeepSeek? Are you utilizing any AIs for programming support? Let us understand in the remarks below.

You can follow my daily project updates on social media. Make sure to sign up for my weekly update newsletter, and follow me on Twitter/X at @DavidGewirtz, on Facebook at Facebook.com/ DavidGewirtz, on Instagram at Instagram.com/ DavidGewirtz, on Bluesky at @DavidGewirtz. com, and on YouTube at YouTube.com/ DavidGewirtzTV.