AMD has introduced the AMD-135M, a compact AI language mannequin. Constructed on Meta’s Llama framework, the AI is particularly designed for personal companies.
In a big transfer, the US-based tech large AMD unveiled the small language mannequin, named AMD-135M, at a latest occasion. As a part of Meta’s “Llama” mannequin, the AMD-135M is predicted to be broadly adopted by non-public enterprises. The mannequin is available in two distinct variations: AMD-Llama-135M and AMD-Llama-135M-code. In keeping with AMD, the AMD-Llama-135M was educated on 670 billion public information tokens, using 4 AMD Intuition MI250s in the course of the coaching course of. The AMD-Llama-135M-code model, however, incorporates a further 20 billion tokens particularly tailor-made for coding duties.
Could be optimized for particular duties
AMD’s small language fashions will be optimized and used for particular duties. Naturally, the AMD-Llama-135M-code shall be primarily used for coding-related capabilities. In keeping with AMD, the brand new language mannequin employs predictive decoding know-how, enabling the fashions to function at excessive pace.
AMD additionally acknowledged that the AMD-135M remains to be in its early phases of improvement. The corporate plans to additional improve the small language mannequin sooner or later, aiming for higher leads to each efficiency and pace. It stays to be seen if AMD, which is striving to determine itself within the synthetic intelligence sector, will obtain the success it seeks with its small language mannequin.
You may additionally like this content material
Comply with us on TWITTER (X) and be immediately knowledgeable concerning the newest developments…
Copy URL
