DeepSeek-Coder-V2: First Open Source Coding Model Beats GPT4-Turbo

Software engineers and LLM enthusiasts, buckle up! DeepSeek-Coder-V2 is here, an open-source AI model that crushes the competition in coding and math tasks. This powerhouse beats out GPT4-Turbo, Claude3-Opus, and more, working with 338 programming languages and understanding complex contexts. DeepSeek-Coder-V2 comes in free 16B and paid 230B versions, both with full access for you to unleash its potential!

Tell us what you think in the comments below!

--------
Chapters:

00:00 - intro
00:45 - DeepSeek V2
00:58 - My AI Coding Setup
01:13 - DeepSeek Updates
01:52 - Benchmarks
07:10 - 338 Programming Languages
09:16 - TESTING!

[fernandoz]

I've been using Deepseek-coder for building some framework in the LLM field and it has been very impresive so far. I cannot imagine what V2 will bring.

[thorvaldspear]

Can't wait for state sponsored LLMs to start introducing code vulnerabilities when the user's location is in a rival country 🤣🤣

[jjhw]

It doesn't need to cost $4M for the chip. you could implement it in an FPGA or on a project chip like Tiny Tapeout, or even a metalized gate array depending on the number of units you need.

[heltengundersen]

Knowing what MooCode is makes me a dinosaur now? They even did it at Stanford!

[SiCSpiT]

I like jailbreaking models and wouldn't mind suffering through rumbles jank to watch how others do it. Especially since anytime I talk about it on youtube my comments disappear.

[yashrid]

Great video. But why you are not running results of code and don't showing up how it handle it from first try?

[Mikoto_]

Codestral 22B and Llama 3 70B are tiny in size compared to gpt-4 t, it would be pathetic if they were better than gpt-4 t

[koctf]

That snake one is obvious wrong.
Generally speaking these models are using the same architecture and training method published by OpenAI. That's why their outputs and capabilities are quite similar.

[jmirodg]

excellent! I'm keen to see the lite model in aider

[Noaman]

I am new php developer which llm i can use Deepseek or which one you recommend ?

[TomNook.]

Wow, you must have just filmed this... Turkey Georgia just finished 2 hours ago!

[techguy]

The wise EEs are all over this stuff. Especially those of us who design both HW and SW.

[Outcast]

Tryed it for coding and it answers the explination in Chinese😂😂😂😂 also it outputs jiberris some times but it could be because of my limited hardware an RTX 3070

[pmarreck]

Which size model did you run and what is its context length?

[amirlilit]

Do you think this model is recommended for VHDL and SystemVerilog coding?

[MrLumatic]

It has great, up to date knowledge. BUT... it's not great in following instructions and calling functions. So it quite often respond in a wrong format, what makes it not-usable on production as Agent Supervisor or for Code Interpreting like with Open Interpreter.
Can't wait to see some fine-tuned version with better instruction following to respond in specific format for Agent Frameworks :)

[GerryPrompt]

Much more impressive than Codestral, any specific reason you use this for your work over other models? Great video! 😊

[hjups]

It's definitely smarter than the previous models, but still fails my simple tests. Although, it did technically get one of the Verilog questions correct if you are purely thinking about inference code (i.e. relying on the compiler to infer a more complex mapping) - perhaps that's due to most of the training set being for FPGAs?
On that same problem, I was able to get it closer to the correct solution when applying the Socratic method, and it came up with a configuration I hadn't considered (which would be more efficient on FPGAs but not ASICs), although it failed to execute on the idea.
For your VHDL test, LLaMA2-7B would get the syntax correct most of the time, so that's not really an effective test. You would have to see how it performed implementing the logic itself, which would probably fail - I noticed it was unable to reason about the parallel nature thinking that the code executed sequentially.
Regardless, the improvement of LLaMA3-70B could make it feasible for local deployment for data privacy reasons when using 4-bit (or 2-bit if it doesn't exhibit a huge loss of quality). Otherwise, I don't see much of a benefit over GPT4.

[remsee]

I tried this on the deepseek website and it sucked seemed like it had a low context length

[VastCNC]

Can you run the code in future videos or generate tests?