DeepSeek 4 Flash local inference engine for Metal
1–10 of 171 posts
Re: DeepSeek 4 Flash local inference engine for Metal
#2[dead]
Re: DeepSeek 4 Flash local inference engine for Metal
#3This 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.
Re: DeepSeek 4 Flash local inference engine for Metal
#4How does it compare to popular local inference engines, e.g. ollama, lm studio, or handrolled llama.cpp? I saw a brief benchmark in the readme but wasn't sure if there was more.
Re: DeepSeek 4 Flash local inference engine for Metal
#5I've been trying deepseek-v4-flash in OpenCode (via OpenRouter) and I'm blown away. It's no Opus, obviously, but it had zero issues with any regular coding task I threw at it. v4-flash is remarkably "good enough" for what I needed. The whole evening of coding cost me $0.52 in API credits.
Re: DeepSeek 4 Flash local inference engine for Metal
#6This 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.
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.
Re: DeepSeek 4 Flash local inference engine for Metal
#7This 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.
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…
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.
Re: DeepSeek 4 Flash local inference engine for Metal
#8This 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.
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…
> 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.
Re: DeepSeek 4 Flash local inference engine for Metal
#9Earlier quoted context omitted.
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…
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.
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.
Re: DeepSeek 4 Flash local inference engine for Metal
#10This 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.
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…
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.