Show HN: Samosa Chat - Run Qwen3.6-35B-A3B Locally on a 16 GB Mac

Post by dwa3592 » Wed, Jul 15, 2026, 2:31 AM UTC

Post by aanet » Wed, Jul 15, 2026, 2:54 AM UTC

Post by prabhanjana_c » Wed, Jul 15, 2026, 5:36 AM UTC

Post by prabhanjana_c » Wed, Jul 15, 2026, 4:26 PM UTC

dwa3592 wrote: post #6

prabhanjana_c wrote: post #5

Very 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.

>>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-princi... maybe give it a try and let me know if your macbook is getting heated while running this?

Post by prabhanjana_c » Thu, Jul 16, 2026, 4:56 AM UTC

dwa3592 wrote: post #8

prabhanjana_c wrote: post #7

Earlier quoted context omitted.

Actually 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.

Thanks for testing. Actually if you open the samosa app; it will show you memory consumption, tokens/sec etc. Let me know what you think.