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

A few words on DS4

antirez.com

Post reply on HN
Print view View original

11–20 of 197 posts

Re: A few words on DS4

#11

Post by kamranjon »

le-mark wrote: post #8
kamranjon wrote: post #6

Just want to mention that I've been pulling down and using DwarfStar locally and it's incredible. I actually have it running on my personal macbook m4 max with 128gb of ram and I am running the server to share it through tailscale with my work laptop and just have pi running there. The long context reasoning is something I haven't even seen in frontier models - I was running at 124k tokens earlier and it was still ju…

Is DS4 dwarf star 4 or deep seek 4?

Just updated! Sorry I meant Dwarf Star - it's the only way I've actually managed to run DeepSeek flash on my local hardware

Re: A few words on DS4

#12

Post by 0xbadcafebee »

I don't see an explanation of why they would make a model-specific inference engine vs just using llamacpp. There are already lots of people working on the llamacpp integration. This is a lot of effort spent on a single model which is likely to become obsolete when a different model comes out that does better. In some discussions, people are now making PRs against both the llamacpp branches and ds4... so it's taking a rare commodity (people investing development time in this model) and fragmenting it

Re: A few words on DS4

#13

Post by xienze »

aiscoming wrote: post #9
xienze wrote: post #7

Earlier quoted context omitted.

I don't want to be a jerk but 31t/s prefill is basically unusable in an agentic situation. A mere 10k in context and you're sitting there for 5+ minutes before the first token is generated.

if it's just the coding agent system prompt and tools, you can cache that

Yeah the problem is that's just the start of the context. There's, you know, all the tool call results and file reads and stuff.

Re: A few words on DS4

#14

Post by kamranjon »

bjconlan wrote: post #2

This is great! I feel the same way about the deepseek v4 architecture for commodity hardware. Also have enjoyed playing with https://huggingface.co/HuggingFaceTB/nanowhale-100m-base (but early days for me understanding this space)

Very cool! I had no idea that HF was doing this - I really love their small model experiments.

Re: A few words on DS4

#15

Post by minimaxir »

A relevant recent tweet from antirez: https://x.com/antirez/status/2054854124848415211

> Gentle reminder on how, in the recent DS4 fiesta, not just me but every other contributor found GPT 5.5 able to help immensely and Opus completely useless.

I've noticed the same for lower level squeezing-as-much-performance-as-possible code work.

Re: A few words on DS4

#16

Post by codedokode »

I thought DeepSeek was closed-weights and proprietary? I wonder how it compares against Western open-weight models. The hugging face page contains the comparison only with proprietary models for some reason.

Re: A few words on DS4

#17

Post by zozbot234 »

0xbadcafebee wrote: post #12

I don't see an explanation of why they would make a model-specific inference engine vs just using llamacpp. There are already lots of people working on the llamacpp integration. This is a lot of effort spent on a single model which is likely to become obsolete when a different model comes out that does better. In some discussions, people are now making PRs against both the llamacpp branches and ds4... so it's taking…

Author has mentioned many times that the llama.cpp maintainers don't want code that's prevalently written by AI with no human revision. If anyone wants to try and get the support upstreamed into that project, they're quite free to do that: the code is MIT licensed.

Re: A few words on DS4

#18

Post by flakiness »

0xbadcafebee wrote: post #12

I don't see an explanation of why they would make a model-specific inference engine vs just using llamacpp. There are already lots of people working on the llamacpp integration. This is a lot of effort spent on a single model which is likely to become obsolete when a different model comes out that does better. In some discussions, people are now making PRs against both the llamacpp branches and ds4... so it's taking…

I believe the assumption is: The code is cheap. The collaboration (eg. upstreaming) is expensive.

Is it true? We'll see, in a few years.

Re: A few words on DS4

#19

Post by zozbot234 »

codedokode wrote: post #16

I thought DeepSeek was closed-weights and proprietary? I wonder how it compares against Western open-weight models. The hugging face page contains the comparison only with proprietary models for some reason.

Nemotron would be a comparable Western open model AIUI.

Re: A few words on DS4

#20

Post by sbinnee »

It is a big thing for sure to have a competitive local agentic model. I've replaced gemini 3 flash preview with DeepSeek v4 flash for all of my personal use cases. Starting from chat app, language learning, and even hobby coding. For coding, I couldn't get decent results no matter which sota latest models I used before. It's not close to Opus or Codex models. It's a flash model and makes mistakes here and there (I just saw `from opentele while import trace`, new Python syntax!)

But I found its tool calling is reliable than other oss models I tried. I assume that it attributes to interleaved thinking. Its reasoning effort is adjusted automatically by queries. I enjoy reading these reasoning traces from open models because you can't see them from proprietary models.

I would love to try DS4 so bad. Well, I don't have a machine for it. I will just stick to openrouter. I wish I can run a competitive oss model on 32GB machine in 3 years.

Post reply on HN