Ask HN: Is anyone using Super Grok Heavy for code?
Curious to hear about experiences from the HN community and how it compares to lower priced options like Claude Max.
Curious to hear about experiences from the HN community and how it compares to lower priced options like Claude Max.
It's currently a distant 12th on https://lmarena.ai/leaderboard/webdev
Only a hair better than grok 3 preview.
Hard pass.
I think its much better than opus. I would describe its code output as boring and to the point, no fluff. O3 Pro is better at abstraction, but grok heavy is better at bug hunting, and only doing exactly as needed. I swapped my openai pro license for grok, its good. Another big advantage is the context window size. Honestly I use these models all day long, and have felt that while sonnet 3.5 was ground breaking, but that anthropic is behind google, openai, and now xai.
For tools, I use repo prompt + grok website. Personally think claude code is overrated, and hand building the context by selecting the files is far better for complicated tasks
How do you find o3-pro for coding? I've also been taking the approach of hand building context and copying and pasting it in for complicated tasks where I want lots of reasoning, like bug/security audits.
I found o1-pro unbelievably good for coding, but when o3-pro was released, I saw the response length in ChatGPT was gimped severely compared to o1-pro, so didn't find it all that useful - it couldn't output long enough responses. I actually cancelled my ChatGPT subscription as it seemed like such a downgrade, though I'll probably try using again via OpenAI's API at some point, so long as the response length isn't capped. I'm tempted to try out Grok 4 Heavy.
o3 pro is really good, but the context is really constrained, so its hard to use and doesnt output enough. This makes it suitable for ideating on good abstractions, but cant really make broad sweeping changes. Grok will output a full file. If you use the o3 pro API, its actually great, but it gets really expensive.
I haven’t used heavy but Grok is excellent at providing one word answers when that’s all you want/need, and I would believe that carries over to heavy
Pretty sure the verbose crap from OpenAI is to jack up output tokens and charge more money.
Interesting theory, and I'd probably believe it if it didn't do it in the free version too.