Tech Meld

Tech-Meld

AI & ML interests

I love ML, i have been learning it since summer of 2023, a lot of things have happened since, from basic knowledge to creation of self-made language models and diffusion models, everything is still in early stages, but work is still done.

Recent Activity

replied to mitkox's post 10 days ago

llama.cpp is 26.8% faster than ollama. I have upgraded both, and using the same settings, I am running the same DeepSeek R1 Distill 1.5B on the same hardware. It's an Apples to Apples comparison. Total duration: llama.cpp 6.85 sec <- 26.8% faster ollama 8.69 sec Breakdown by phase: Model loading llama.cpp 241 ms <- 2x faster ollama 553 ms Prompt processing llama.cpp 416.04 tokens/s with an eval time 45.67 ms <- 10x faster ollama 42.17 tokens/s with an eval time of 498 ms Token generation llama.cpp 137.79 tokens/s with an eval time 6.62 sec <- 13% faster ollama 122.07 tokens/s with an eval time 7.64 sec llama.cpp is LLM inference in C/C++; ollama adds abstraction layers and marketing. Make sure you own your AI. AI in the cloud is not aligned with you; it's aligned with the company that owns it.

new activity 10 days ago

HuggingFaceTB/SmolVLM-256M-Instruct:There is an issue with AutoProcessor

replied to mitkox's post 11 days ago

View all activity

Organizations

None yet

Tech-Meld's activity

replied to mitkox's post 10 days ago

Thanks for pointing that!

New activity in HuggingFaceTB/SmolVLM-256M-Instruct 10 days ago

There is an issue with AutoProcessor

#6 opened 10 days ago by

Tech-Meld

replied to mitkox's post 11 days ago

how did you get Deepseek R1 Qwen 1.5B running with llama.cpp? i made a small app with llama.cpp bindings for python but when trying to load this new model i get an error saying this model is not supported, please help!

New activity in yuntian-deng/o1 24 days ago