servius/dotfiles
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

feat: Added minstral and rag embedding model for aichat

Browse Source
This commit is contained in:
uttarayan21
2024-11-29 15:13:13 +05:30
parent 9b79427ecf
commit 4b2b3f9858
2 changed files with 68 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
common/home.nix
View File

@@ -356,6 +356,22 @@ in {
}
];
}
{
type = "openai-compatible";
name = "ollama";
api_base = "https://llama.darksailor.dev/api/ollama/v1";
api_key_cmd = "op item get llama-api --fields label=credential --reveal";
models = [
{
name = "RobinBially/nomic-embed-text-8k";
type = "embedding";
default_chunk_size = 8000;
}
{
name = "codellama:minstral";
}
];
}
{
type = "openai";
name = "openai";
@@ -371,6 +387,7 @@ in {
];
}
];
rag_embedding_model = "ollama:RobinBially/nomic-embed-text-8k";
model = "openai:gpt-4o";
};
};

57
nixos/mirai/services.nix
View File

@@ -140,6 +140,15 @@
configureRedis = true;
https = true;
};
ollama = {
enable = true;
loadModels = ["RobinBially/nomic-embed-text-8k" "minstral"];
port = 11434;
host = "0.0.0.0";
environmentVariables = {
OLLAMA_ORIGINS = "*";
};
};
llama-cpp = {
enable = true;
host = "127.0.0.1";
@@ -149,12 +158,6 @@
sha256 = "61834b88c1a1ce5c277028a98c4a0c94a564210290992a7ba301bbef96ef8eba";
url = "https://huggingface.co/bartowski/Qwen2.5.1-Coder-7B-Instruct-GGUF/resolve/main/Qwen2.5.1-Coder-7B-Instruct-Q8_0.gguf?download=true";
};
# extraFlags = ["--" "--in-prefix" "<|im_start|>" "--in-suffix" "<|im_end|>"];
# model = builtins.fetchurl {
# name = "mistral-7b-claude-chat";
# sha256 = "03458d74d3e6ed650d67e7800492354e5a8a33aaaeabc80c484e28766814085a";
# url = "https://huggingface.co/TheBloke/Mistral-7B-Claude-Chat-GGUF/resolve/main/mistral-7b-claude-chat.Q8_0.gguf?download=true";
# };
};
nginx.virtualHosts."${config.services.nextcloud.hostName}".listen = [
{
@@ -180,6 +183,11 @@
reverse_proxy localhost:3000
}
handle /api/ollama/* {
uri strip_prefix /api/ollama
reverse_proxy localhost:11434
}
handle {
forward_auth localhost:5555 {
uri /api/authz/forward-auth
@@ -204,4 +212,41 @@
'';
};
};
# containers.llama = {
# autoStart = true;
# privateNetwork = true;
# hostAddress = "192.168.100.10";
# localAddress = "192.168.100.11";
# hostAddress6 = "fc00::1";
# localAddress6 = "fc00::2";
# config = {
# config,
# pkgs,
# libs,
# ...
# }: {
# system.stateVersion = "24.11";
# networking = {
# firewall = {
# enable = true;
# allowedTCPPorts = [4000];
# };
# # Use systemd-resolved inside the container
# # Workaround for bug https://github.com/NixOS/nixpkgs/issues/162686
# useHostResolvConf = lib.mkForce false;
# };
# services.resolved.enable = true;
# services.llama-cpp = {
# enable = true;
# host = "127.0.0.1";
# port = 4000;
# model = builtins.fetchurl {
# name = "qwen_2.5.1_coder_7b_instruct_gguf";
# sha256 = "61834b88c1a1ce5c277028a98c4a0c94a564210290992a7ba301bbef96ef8eba";
# url = "https://huggingface.co/bartowski/Qwen2.5.1-Coder-7B-Instruct-GGUF/resolve/main/Qwen2.5.1-Coder-7B-Instruct-Q8_0.gguf?download=true";
# };
# };
# };
# };
}