Performance of llama.cpp on Apple Silicon M-series · ggerganov/llama.cpp · Discussion #4167
# GPT-4 Generated Code import pandas as pd import numpy as np import matplotlib.pyplot as plt # Creating DataFrame from the provided data data = { "Chip": ["M1", "M1", "M1 Pro", "M1 Pro", "M1 Max", "M1 Max", "M1 Ultra", "M2", "M2 Pro", "M2 Pro", "M2 Max", "M2 Max", "M2 Ultra", "M2 Ultra", "M3", "M3 Pro", "M3 Pro", "M3 Max"], "BW (GB/s)": [68, 68, 200, 200, 400, 400, 800, 100, 200, 200, 400, 400, 8
1
リリース、障害情報などのサービスのお知らせ
最新の人気エントリーの配信
処理を実行中です
j次のブックマーク
k前のブックマーク
lあとで読む
eコメント一覧を開く
oページを開く
prompt is too long · ggerganov/llama.cpp · Discussion #5786
creating gguf model from lora adapter · ggerganov/llama.cpp · Discussion #5360
Llama 2 70b Chat not working on M1 Macs when using Metal · Issue #2429 · ggerganov/llama.cpp
converting LORA to ggml to gguf · Issue #3953 · ggerganov/llama.cpp
Issues of converting safetensors to gguf with convert.py · Issue #5278 · ggerganov/llama.cpp
how to convert qlora trained model into a GGUF model? · ggerganov/llama.cpp · Discussion #3489
Benchmarking: What types of performance does LLMs care about? · ggerganov/llama.cpp · Discussion #2038
llama : Metal inference by ggerganov · Pull Request #1642 · ggerganov/llama.cpp
Add GPU support to ggml · ggerganov/llama.cpp · Discussion #915
