ggml/llama.cpp releases: b11908

TL;DR

The ggml/llama.cpp project updated its codebase by moving the `no_alloc` parameter out of `llama_hparams`, aiming to improve memory management and efficiency. These changes are part of ongoing optimizations, assisted by high-performance computing resources, highlighting efforts to enhance the performance of language models.

Detailed Summary

The ggml/llama.cpp project released updates that moved the `no_alloc` attribute out of `llama_hparams`, aiming to improve memory management in language models. These changes were assisted by dsh using two DGX Spark systems with RDMA and a throughput of 2. The broader impact includes potential enhancements in model efficiency and performance, though specific outcomes are not detailed in this update.

Key Points

• • llama : move no_alloc out of llama_hparams
• • cont : clean-up
• • cont : clean-up