Ollama vs llama.cpp

Cluster focuses on comparisons between Ollama and llama.cpp for running local LLMs, debates on their benefits, compatibility, use cases, and alternatives like llamafile or oobabooga.

➡️ Stable 0.9x AI & Machine Learning
5,896
Comments
11
Years Active
5
Top Authors
#7543
Topic ID

Activity Over Time

2008
1
2010
1
2017
1
2019
1
2020
1
2021
2
2022
12
2023
2,136
2024
1,861
2025
1,787
2026
95

Keywords

HuggingFace adf.tfId LLM OK Lama.cpp LangChain GPU UI CUDA AMD llama cpp models run model studio local hosting use better chat

Sample Comments

buyucu Aug 12, 2025 View on HN

ollama is still using llama.cpp. they are just denying that they are :)

nunodonato Jan 7, 2026 View on HN

don't use ollama. llama.cpp is better because ollama has an outdated llama.cpp

thehamkercat Dec 21, 2025 View on HN

That's why i suggested using llama.cpp in my other comment.

eclectic29 Feb 9, 2024 View on HN

What's the use case of Ollama? Why should I not use llama.cpp directly?

v3ss0n Dec 21, 2023 View on HN

try ollama , only needs about 4GB it uses llmcpp

killingtime74 Dec 21, 2023 View on HN

What's the easiest way to adapt this to local LLMs like Ollama or Lama.cpp

pryelluw Dec 25, 2025 View on HN

Any plans to support local models through llama.cpp or similar?

Rhubarrbb Aug 8, 2025 View on HN

Does it work with local LLMs like through Ollama or llama.cpp?

iruoy Apr 13, 2023 View on HN

Will Llama.cpp give the same results as Llama? And how is it so much easier to run?

xrd Jul 18, 2023 View on HN

Does anyone know if this works with llama.cpp?