Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

GitHub Copilot is now available for free

github.com

Post reply on HN
Print view View original

331–340 of 412 posts

Re: GitHub Copilot is now available for free

#331

Post by weikju »

int_19h wrote: post #221
rvz wrote: post #63

Earlier quoted context omitted.

All of the above. It is even worse than crypto. The AI proponents that support this have no serious solution to the mass displacement of jobs thanks to AI. They actually don't mention any alternative solutions and instead scream about nonsense such as UBI which has never worked at a large sustainable scale. > Society needs to slow down and find alternative, more sustainable solutions. AI is aligned with short-term ec…

If robots do all the work (which is ultimately what all the lost jobs are about), why would UBI be unsustainable at scale?

Because the rich and powerful people who will reap the most benefit from all the automation will not redistribute the wealth to the now-useless ex-labor force.

Re: GitHub Copilot is now available for free

#332

Post by scosman »

thot_experiment wrote: post #241
scosman wrote: post #215

Earlier quoted context omitted.

yeah, Sonnet goes past that. 300+ line changes in 20 seconds. You have to review it, but generally it's right. It's infinitely faster than the time to look at docs and do it myself. Sure it's busywork. But it's a lot of busywork very fast.

Well it's definitely not infinitely faster since you're having to review it, but we're talking about the delta between Sonnet and Qwen/Mistral/Llama or whatever, not doing it manually. I'm really curious what your problem domain is, like specifically what sort of code are you asking it to change and what changes are you asking for. I just gave o1 and Sonnet a total layup question (optimization that had a huge win sim…

Quick example: create a config class, allowing the caller get and set a variety of config vars. It should be easy to add new vars. Persist any set value to a yaml file. Each var should be described by a name, type, env var fallback value (optional) and default value if the set value and env var are null (optional). The API should be typed and allow int, string, float and lists. Add comprehensive tests.

Obviously nothing complicated but it takes non-zero time. It did it in one shot in about 30s. Didn’t have to look at and docs (I don’t have the yaml lib memorized). Got the python typing right which would have been a bit of a pain. A lot faster than doing it myself, even with reviews. Tests were solid so I could tell it worked.

The filter example you give seems like they should have aced it. Not sure what went wrong but it has easily done work like that for me. I usually am half way through typing the method name when the rest autocompletes. Are you using a tool with good context management like cursor?

Re: GitHub Copilot is now available for free

#333

Post by zxcq544 »

euoia wrote: post #289
solarkraft wrote: post #281

I have already replaced Copilot with continue.dev+qwen2.5-coder:1.5b on Ollama and don't see myself coming back. For the last year Copilot completions have been slow, unreliable and just bad , coming in at random moments, messing up syntax in dumb ways, being slow and sometimes not showing up at all. It has been painfully bad even though it used to be good. Maybe it's since they switched from Codex to the general GPT…

Do you have a guide for how to set this up? I am also pretty dissatisfied with Copilot completions.

Use Continue https://ollama.com/blog/continue-code-assistant

Re: GitHub Copilot is now available for free

#334

Post by x187463 »

vbezhenar wrote: post #97
__jl__ wrote: post #3

Has anyone compared the updated Copilot with Cursor? The main updates I am wondering about are model selection and multi-file edits. I used copilot before these features, changed to Cursor and now I am wondering how much Copilot has closed the gap.

I'm using Copilot daily and I didn't find any improvements over last year. I think they introduced something called Copilot Edits which works kind of like aider judging from screenshots, but it's experimental, so I didn't try it. It's basically glorified autocomplete and that's about it. At least I didn't discover anything new.

The glorified autocomplete only applies to the completions. Granted, that's often the primary interaction, but it isn't the whole story. Using chat+edits can enable copilot to make changes across entire file(s). I still haven't found it perfectly reliable for large scale edits, but it has proved useful for handling the sort of busywork that may occur when refactoring. I'd love to hear stories where copilot actually made meaningful contributions to unique/interesting projects. The demo's often showcase making mundane modifications to yet-another-website or spawning a project into existence, which is neither very compelling nor something I do often.

Re: GitHub Copilot is now available for free

#336

Post by notsylver »

solarkraft wrote: post #281

I have already replaced Copilot with continue.dev+qwen2.5-coder:1.5b on Ollama and don't see myself coming back. For the last year Copilot completions have been slow, unreliable and just bad , coming in at random moments, messing up syntax in dumb ways, being slow and sometimes not showing up at all. It has been painfully bad even though it used to be good. Maybe it's since they switched from Codex to the general GPT…

I've had the opposite experience - I tried continue.dev and for me it doesn't come close to copilot. Especially with copilot chat having o1-preview and Sonnet 3.5 for so cheap that I single handedly might bankrupt microsoft (we can hope), but I tried it before that was availlable and the inline completions were laughably bad in comparison.

I used the recommended models and couldn't figure it out, I assume I did something wrong but I followed the docs and triple checked everything. It'd be nice to use the GPU I have locally for faster completions/privacy, I just haven't found a way to do that.

Re: GitHub Copilot is now available for free

#337

Post by solarkraft »

euoia wrote: post #289
solarkraft wrote: post #281

I have already replaced Copilot with continue.dev+qwen2.5-coder:1.5b on Ollama and don't see myself coming back. For the last year Copilot completions have been slow, unreliable and just bad , coming in at random moments, messing up syntax in dumb ways, being slow and sometimes not showing up at all. It has been painfully bad even though it used to be good. Maybe it's since they switched from Codex to the general GPT…

Do you have a guide for how to set this up? I am also pretty dissatisfied with Copilot completions.

Here you go: https://docs.continue.dev/autocomplete/model-setup

The sibling comment also describes the process for chat, which I personally don’t care about.

Re: GitHub Copilot is now available for free

#338

Post by darthwalsh »

Gigachad wrote: post #139
sleepybrett wrote: post #73

Earlier quoted context omitted.

... and they aren't concerned with what microsoft is doing with it?

Microsoft already has access to all of your code on Github.

Many big companies are using GitHub Enterprise Server or some other self-hosted version control.

GitHub Copilot works fine with these; just authenticate against github.com first

Re: GitHub Copilot is now available for free

#339

Post by indigodaddy »

int_19h wrote: post #218
indigodaddy wrote: post #149

Earlier quoted context omitted.

with how much ram though?

For Qwen-32B, 32Gb should be quite sufficient to run it at 4-bit quantization with full context.

Thanks, I need to figure out something to run on M4 Mac Mini standard 16G

Re: GitHub Copilot is now available for free

#340

Post by fortyseven »

notsylver wrote: post #336
solarkraft wrote: post #281

I have already replaced Copilot with continue.dev+qwen2.5-coder:1.5b on Ollama and don't see myself coming back. For the last year Copilot completions have been slow, unreliable and just bad , coming in at random moments, messing up syntax in dumb ways, being slow and sometimes not showing up at all. It has been painfully bad even though it used to be good. Maybe it's since they switched from Codex to the general GPT…

I've had the opposite experience - I tried continue.dev and for me it doesn't come close to copilot. Especially with copilot chat having o1-preview and Sonnet 3.5 for so cheap that I single handedly might bankrupt microsoft (we can hope), but I tried it before that was availlable and the inline completions were laughably bad in comparison. I used the recommended models and couldn't figure it out, I assume I did somet…

The last couple times I tried "continue" it felt like "Step 1" in someone's business plan; bulky and seconds away from converting into a paid subscription model.

Additionally, I've tried a bunch of these (even the same models, etc) and they've all sucked compared to Copilot. And believe me, I want that local-hosted sweetness. Not sure what I'm doing wrong when others are so excited by it.

Post reply on HN