header-langage
简体中文
繁體中文
English
Tiếng Việt
한국어
日本語
ภาษาไทย
Türkçe
Scan to Download the APP
Partnered With:

Kimi K3 can now run locally on Mac, but regular users still cannot afford it

According to Dynamic Beating monitoring, the open-source AI tool company Unsloth has released the 1-bit GGUF version of Kimi K3, compressing the model from 1.56TB to 594GB. Users can run it locally on a Mac through Unsloth Studio or llama.cpp.

However, this version requires at least about 610GB of memory. Following the official recommendation, the top-of-the-line 512GB Mac Studio needs to be paired with a 128GB memory device. Insufficient memory can also be supplemented by loading from a hard drive, but this will result in a significant speed reduction.

The community also offers 350GB and 451GB MLX trimmed versions that can fit into a single 512GB Mac Studio. However, they have removed 73% to 80% of the experts, with observed speeds of approximately 0.14 to 0.20 tokens per second, showing a noticeable decrease in performance.

举报 Correction/Report
RelatedRecommended
Foreigners said they were more shocked than the day the Soviet satellite was launched into orbit.
Citigroup and Bank of America Securities believe that while the Kimi K3 reduces inference costs, it may paradoxically stimulate more invocations, longer contexts, and agent-based tasks through the "Jevons Paradox," thereby driving up demand for GPU, HBM, DDR5, eSSD, and network infrastructure.
Low-Cost Open-Weight Model Challenges AI Valuation Narrative
Correction/Report
Submit
Add Library
Visible to myself only
Public
Save
Choose Library
Add Library
Cancel
Finish