Merge branch 'master' into concedo_experimental

# Conflicts:
#	.dockerignore
#	.github/workflows/build.yml
#	CMakeLists.txt
#	Makefile
#	README.md
#	flake.lock
#	flake.nix
#	tests/CMakeLists.txt

scripts/convert-gg.sh

@@ -0,0 +1,26 @@
+#!/bin/bash
+
+set -e
+
+# LLaMA v1
+python3 convert.py ../llama1/7B  --outfile models/llama-7b/ggml-model-f16.gguf  --outtype f16
+python3 convert.py ../llama1/13B --outfile models/llama-13b/ggml-model-f16.gguf --outtype f16
+python3 convert.py ../llama1/30B --outfile models/llama-30b/ggml-model-f16.gguf --outtype f16
+python3 convert.py ../llama1/65B --outfile models/llama-65b/ggml-model-f16.gguf --outtype f16
+
+# LLaMA v2
+python3 convert.py ../llama2/llama-2-7b  --outfile models/llama-7b-v2/ggml-model-f16.gguf  --outtype f16
+python3 convert.py ../llama2/llama-2-13b --outfile models/llama-13b-v2/ggml-model-f16.gguf --outtype f16
+python3 convert.py ../llama2/llama-2-70b --outfile models/llama-70b-v2/ggml-model-f16.gguf --outtype f16
+
+# Code Llama
+python3 convert.py ../codellama/CodeLlama-7b/  --outfile models/codellama-7b/ggml-model-f16.gguf  --outtype f16
+python3 convert.py ../codellama/CodeLlama-13b/ --outfile models/codellama-13b/ggml-model-f16.gguf --outtype f16
+python3 convert.py ../codellama/CodeLlama-34b/ --outfile models/codellama-34b/ggml-model-f16.gguf --outtype f16
+
+# Falcon
+python3 convert-falcon-hf-to-gguf.py ../falcon/falcon-7b  1
+mv -v ../falcon/falcon-7b/ggml-model-f16.gguf models/falcon-7b/ggml-model-f16.gguf
+
+python3 convert-falcon-hf-to-gguf.py ../falcon/falcon-40b 1
+mv -v ../falcon/falcon-40b/ggml-model-f16.gguf models/falcon-40b/ggml-model-f16.gguf

scripts/qnt-all.sh

@@ -0,0 +1,29 @@
+#!/bin/bash
+
+qnt=(q8_0 q6_k q5_k q5_1 q5_0 q4_k q4_1 q4_0 q3_k q2_k)
+args=""
+
+if [ -z "$1" ]; then
+    echo "usage: $0 <model> [qnt] [args]"
+    echo "default: $0 <model> \"${qnt[@]}\" \"${args}\""
+    exit 1
+fi
+
+if [ ! -z "$2" ]; then
+    qnt=($2)
+fi
+
+if [ ! -z "$3" ]; then
+    args="$3"
+fi
+
+model="$1"
+out="../tmp/results-${model}"
+
+set -e
+
+mkdir -p ${out}
+
+for q in ${qnt[@]}; do
+    time ./bin/quantize ../models/${model}/ggml-model-f16.gguf ../models/${model}/ggml-model-${q}.gguf ${q} 2>&1 ${args} | tee ${out}/qnt-${q}.txt
+done

scripts/run-all-perf.sh

@@ -0,0 +1,33 @@
+#!/bin/bash
+
+qnt=(f16 q8_0 q6_k q5_k q5_1 q5_0 q4_k q4_1 q4_0 q3_k q2_k)
+args="-ngl 999 -n 64 -p 512"
+
+if [ -z "$1" ]; then
+    echo "usage: $0 <model> [qnt] [args]"
+    echo "default: $0 <model> \"${qnt[@]}\" \"${args}\""
+    exit 1
+fi
+
+if [ ! -z "$2" ]; then
+    qnt=($2)
+fi
+
+if [ ! -z "$3" ]; then
+    args="$3"
+fi
+
+model="$1"
+out="../tmp/results-${model}"
+
+set -e
+
+mkdir -p ${out}
+
+mstr=""
+
+for q in ${qnt[@]}; do
+    mstr="${mstr} -m ../models/${model}/ggml-model-${q}.gguf"
+done
+
+./bin/llama-bench ${mstr} ${args} 2> /dev/null

scripts/run-all-ppl.sh

@@ -0,0 +1,29 @@
+#!/bin/bash
+
+qnt=(f16 q8_0 q6_k q5_k q5_1 q5_0 q4_k q4_1 q4_0 q3_k q2_k)
+args="-ngl 999 -t 8"
+
+if [ -z "$1" ]; then
+    echo "usage: $0 <model> [qnt] [args]"
+    echo "default: $0 <model> \"${qnt[@]}\" \"${args}\""
+    exit 1
+fi
+
+if [ ! -z "$2" ]; then
+    qnt=($2)
+fi
+
+if [ ! -z "$3" ]; then
+    args="$3"
+fi
+
+set -e
+
+model="$1"
+out="../tmp/results-${model}"
+
+mkdir -p ${out}
+
+for q in ${qnt[@]}; do
+    time ./bin/perplexity -m ../models/${model}/ggml-model-f16.gguf -f ./wiki.test.raw ${args} 2>&1 | tee ${out}/ppl-${q}.txt
+done