DeepSeek 4 Flash local inference engine for Metal

(github.com)

76 points | by tamnd 3 hours ago ago

22 comments

  • antirez 21 minutes ago

    A random, funny, interesting and telling data point: my MacBook M3 Max while DS4 is generating tokens at full speed peaks 50W of energy usage...

    • minimaxir 15 minutes ago

      "Data centers for LLMs are technically more energy efficient per-user than self-hosting LLM models due to economies-of-scale" is a data point the internet isn't ready for.

      • fgfarben 8 minutes ago

        Because it isn't correct, right?

        Because it isn't correct, right?

      • Onavo 8 minutes ago

        There's a bunch of companies doing garage GPU datacenters now. Probably can act as a heat source during winter too if you have a heat pump.

    • bertili 15 minutes ago

      equals 2 or 3 human brains in power usage. Amazing work!

      • antirez 7 minutes ago

        True quantitatively, not qualitatively. DeepSeek V4 is not capable of doing what a human brain can do, of course, but for the tasks it can do, it can do it at a speed which is completely impossible for a human, so comparing the two requires some normalization for speed.

    • Hamuko 12 minutes ago

      I think I’ve seen about 60 watt total system whenever I’ve used a local model on a MacBook Pro or a Mac Studio. Baseline for the Mac Studio is like 10 W and like 6 W for the MacBook Pro.

  • maherbeg 2 hours ago

    This is so sick. I'm really curious to see what focused effort on optimizing a single open source model can look like over many months. Not only on the inference serving side, but also on the harness optimization side and building custom workflows to narrow the gap between things frontier models can infer and deduce and what open source models natively lack due to size, training etc.

    • dakolli an hour ago

      There will always be a huge gap between frontier models and open source models (unless you're very rich). This whole industry makes no sense, everyone is ignoring the unit economics. It cost 20k a month to running Kimi 2.6 at decent tok/ps, to sell those tokens at a profit you'd need your hardware costs to be less 1k a month.

      Everyone who's betting their competency on the generosity of billionaires selling tokens for 1/10-1/20th of the cost, or a delusional future where capable OS models fit on consumer grade hardware are actually cooked.

      • bensyverson an hour ago

        If you looked at a graph of GPU power in consumer hardware and model capability per billion parameters over time, it seems inevitable that in the next few years a "good enough" model will run on entry-level hardware.

        Of course there will always be larger flagship models, but if you can count on decent on-device inference, it materially changes what you can build.

        • physicsguy 44 minutes ago

          It also massively changes the value economics of the frontier models. In a lot of cases, you really don't need a general purpose intelligence model too.

        • dakolli 37 minutes ago

          No offense, this is a crazy delusional statement.

          • afro88 33 minutes ago

            No offense, this is a crazy worthless contribution to the discussion.

            Why?

      • liuliu 44 minutes ago

        I am not sure where this comment is from (possibly without looking at this project?). This project is running quasi-frontier model at reasonable tps (~30) with reasonable prefill performance (~500tps) with a high-end laptop. People simply project what they see from this project to what you optimistically can expect.

        You can argue whether the projection is too optimistic or not, but this project definitely made me a little bit optimistic on that end.

      • amunozo 26 minutes ago

        Most tasks do not require frontier models, so as long as these models cover 95-99 per cent of the tasks, closed frontier models can be left for niche and specialized cases that are harder.

      • otabdeveloper4 44 minutes ago

        > a delusional future where capable OS models fit on consumer grade hardware

        48 gb is enough for a capable LLM.

        Doing that on consumer grade hardware is entirely possible. The bottleneck is CUDA and other intellectual property moats.

  • amunozo 28 minutes ago

    I am curious about it producing less tokens except for the max mode. I love DeepSeek V4 Flash and I use it extensively, it's so cheap I can use it all day and still not use all my 10$ OpenCode Go subscription. I use it always in max mode because of this, but now I wonder whether I should rather use high.

    • syntaxing a minute ago

      How has opencode go been for you? Worth changing over from Claude pro?

    • unshavedyak 8 minutes ago

      What do you use it for? I tend to just stick to SOTA (Claude 4.7 Max thinking), and put up with the slow req/response. I'm not sure what type of work i'd trust a less thinking model, as my intuition is built around what Claude vSOTA Max can handle.

      Nonetheless eventually i want to build an at-home system. I imagine some smaller local model could handle metadata assignment quite well.

  • happyPersonR 10 minutes ago

    So just gonna ask a question, probably will get downvoted

    I know this is flash, but….

    But other than this guy, did our whole society seriously never flamegraph this stuff before we started requesting nuclear reactors colocated at data centers and like more than 10% of gdp?

    Someone needs to answer because this isn’t even a m4 or m5… WHAT THE FUCK

    Sidenote: shout out antirez love my redis :)

    • AlotOfReading 2 minutes ago

      This is built atop a tower of stuff people built with profiling and performance-oriented design.

      That said, I've found that most corporate environments are unintentionally hostile to this kind of optimization work. It's hard to justify until the work is already done. That means you often need people with the skills, means, and motivation to do this that are outside normal corporate constraints. There aren't many of those.

    • liuliu 9 minutes ago

      DSv4 generates much faster on NVIDIA class hardware. It is just a very efficient model.