4 ms·
Show HN: Samosa Chat - Run Qwen3.6-35B-A3B Locally on a 16 GB Mac
- daef 3mo agoany way to run samosa in opencode?
- MrGrinchh 3mo agoThis is pretty sick. 7tokens/second is not fast but still seems useful.
- aanet 3mo ago+1 for pairing Samosa with Chāt :D :D
- dwa3592 3mo agoi was pleasantly surprised to find out it hadn't been done until now.
- prabhanjana_c 3mo agoVery interesting. I am always curious on llama.cpp and vllm. All the best for samosa inference engine. One reason I am not running local models on my Mac now a days, is My Mac book pro is getting heated more. Otherwise would love to run couple (at least 1 to 3) of local models continuously.
- dwa3592 3mo ago>>One reason I am not running local models on my Mac now a days, is My Mac book pro is getting heated more Samosa was optimized for keeping mac's temp in check. This is one of the important reasons for creating this - no excessive wear/tear to the machine. https://github.com/deepanwadhwa/samosa-chat#the-three-principles https://github.com/deepanwadhwa/samosa-chat#the-three-princi... maybe give it a try and let me know if your macbook is getting heated while running this?
- prabhanjana_c 3mo agoActually quite cool. My mac is not getting heated. will check more. Memory pressure is very low. I see it (memory used) just increased from 22GB to 27Gb(on 48GB pro), while using somosa command(again a very basic test). Not really measured tokens/s, but I see it's reasonable.
- dwa3592 3mo agoThanks for testing. Actually if you open the samosa app; it will show you memory consumption, tokens/sec etc. Let me know what you think.
- prabhanjana_c 3mo agoGreat, yes with samosa app can see details, it is 6.52tok/sec, 4.05GB memory. Use case, I have been waiting was to run some agents in my Mac continuously with local models, but without my Mac getting heated. Seems I can experiment with samosa inference engine.