[HN Gopher] Stable Code 3B: Coding on the Edge
       ___________________________________________________________________
        
       Stable Code 3B: Coding on the Edge
        
       Author : egnehots
       Score  : 67 points
       Date   : 2024-01-16 21:40 UTC (1 hours ago)
        
 (HTM) web link (stability.ai)
 (TXT) w3m dump (stability.ai)
        
       | keyle wrote:
       | That is fantastic. I'm building a small macOS SwiftUI client with
       | llama cpp built in, no server-client model, and it's already so
       | useful with models like openhermes chat 7B, and fast.
       | 
       | If this opens it to smaller laptops, wow!
       | 
       | We truly live in crazy time. The rate of improvement in this
       | field is off the walls.
        
         | turnsout wrote:
         | That sounds awesome! Can you share any details about how you're
         | working with llama cpp? Is it just via the Swift <> C bridge?
         | I've toyed with the idea of doing this, and wonder if you have
         | any pointers before I get started.
        
         | joshmarlow wrote:
         | Not sure if this is where your head is, but I think there's a
         | lot of value in integrating LLMs directly into complex
         | software. Jira, Salesforce, maybe K8s - should all have an
         | integrated LLMs that can walk you through how to perform a
         | nuanced task in the software.
        
           | dpacmittal wrote:
           | Why would the LLM walk you through and not just do the
           | nuanced task on its own?
        
             | pennomi wrote:
             | I assume the human maintains some of the necessary context
             | in their meat memory.
        
       | artninja1988 wrote:
       | Given the complete failure of the first stable lm, I'm interested
       | to try this one out. Haven't really seen a small language model,
       | except mixtral 7b that's really useful for much.
       | 
       | I also hope stability comes out with a competitor to the new
       | midjourney and dalle models! That's what put them on the map in
       | the first place
        
         | brianjking wrote:
         | All of the Mistral versions have been excellent, including the
         | OpenHermes versions. I encourage you to check out Phi-2 as
         | well, it's the only 3b model I've found really quite
         | interesting outside of Replit's code model built into Replit
         | Core.
        
         | tarruda wrote:
         | Deepseek coder 6.7B is very useful for coding and can run in
         | consumer GPUs.
         | 
         | I use the 6bit GGUF quantized version on a laptop RTX 3070
        
       | mchiang wrote:
       | It's amazing to see more smaller models being released. This
       | creates opportunities for more developers to run it on their
       | local computers, and makes it easier to fine-tune for specific
       | needs.
        
         | brcmthrowaway wrote:
         | Has anyone tried starting with a smaller modeling, then RLing
         | until it improves to the bigger model?
        
       | lfkdev wrote:
       | How is this compared to the current GitHub Copilot?
        
         | brianjking wrote:
         | A 3B tiny model is not going to compare to GitHub copilot.
         | However, there are plenty of nice 7B models that are excellent
         | at code and I encourage you to try them out.
        
         | londons_explore wrote:
         | If you just want to get stuff done, use the best tools like a
         | Milwaukee Drill - and right now, thats copilot/gpt-4.
         | 
         | If you don't want to be tied to a company and like opensource,
         | feel free to connect a toy motor to an AA battery to drill your
         | holes... Or to use Llama/Stable Code 3B.
        
           | mistercheph wrote:
           | it's going to be real hard to pry the carburetors out of this
           | guy's cold dead hands!
        
       | photon_collider wrote:
       | How reliable are these benchmarks?
        
         | ilaksh wrote:
         | I think the trick is that they are just comparing to other tiny
         | models.
         | 
         | None of the little models, including this one, are comparable
         | to the performance of the larger models for any significant
         | coding problem.
         | 
         | I think what these are useful for is mostly giving people hints
         | inside of a code editor. Occasionally filling in the blank.
        
       | knicholes wrote:
       | I've got a machine with 4 3090s-- Anyone know which model would
       | perform the best for programming? It's great this can run on a
       | machine w/out a graphics card and is only 3B params, but I have
       | the hardware. Might as well use it.
        
         | SushiHippie wrote:
         | Here is a leader board of some models
         | 
         | https://huggingface.co/spaces/mike-ravkine/can-ai-code-resul...
         | 
         | Don't know how biased this leaderboard is, but I guess you
         | could just give some of them a try and see for yourself.
        
       | tarruda wrote:
       | Note that they don't compare with deepseek coder 6.7b, which is
       | vastly superior to much bigger coding models. Surpassing
       | codellama 7b is not that big of a deal today.
       | 
       | The most impressive thing about these results is how good the
       | 1.3B deepseek coder is.
        
       | connorgutman wrote:
       | FYI: This model is already available on Ollama.
        
       | kleiba wrote:
       | It's quite amazing - I often find that I read quite positive
       | comments towards LLM tools for coding. Yet, an "Ask HN" I posted
       | a while ago (and which admittedly didn't gain much traction)
       | seemed to mirror mostly negative/pessimistic responses.
       | 
       | https://news.ycombinator.com/item?id=38803836
       | 
       | Was it just that my submission didn't find enough / more balanced
       | commenters?
        
         | simonw wrote:
         | You only got comments from six people so yeah, definitely not
         | representative.
        
       | akulbe wrote:
       | I just tried this model with Koboldcpp on my LLM box. I got
       | gibberish back.
       | 
       | My prompt - "please show me how to write a web scraper in Python"
       | 
       | The response?
       | 
       | <blockquote> I've written my first ever python script about 5
       | months ago and I really don't remember anything except for the
       | fact that I used Selenium in order to scrape websites (in this
       | case, Google). So you can probably just copy/paste all of these
       | lines from your own Python code which contains logic to determine
       | what value should be returned when called by another piece of
       | software or program. </blockquote>
        
         | endofreach wrote:
         | This doesn't seem like gibberish though?
        
         | connorgutman wrote:
         | Same thing with Ollama.
        
         | Tiberium wrote:
         | But it's a code completion model, not a chat/instruct one.
        
         | SushiHippie wrote:
         | It's very likely a "completion model" and not instruct/chat
         | fine-tuned.
         | 
         | So you'd need to prompt it through comments or by starting with
         | a function name, basically the same as one would prompt GitHub
         | copilot.
         | 
         | e.g.                 # the following code implements a
         | webscraper in python       class WebScraper:
         | 
         | (I didn't try this, and I'm not good at prompting, but
         | something along the lines of this example should yield better
         | results)
        
         | MrNeon wrote:
         | It is weird that it is not mentioned in the model card but I'm
         | pretty sure it is a completion model, not tuned as an instruct
         | model.
         | 
         | edit: the webpage does call it "Stable Code Completion"
        
       ___________________________________________________________________
       (page generated 2024-01-16 23:00 UTC)