TheCrazyInsanity/nixos-configv3
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

idk

Browse Source
This commit is contained in:
TheCrazyInsanity
2026-04-10 16:09:45 -04:00
parent ac5425e5b1
commit a1fe50ee54
2 changed files with 6 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
global/default.nix
View File

@@ -32,13 +32,13 @@
blasSupport = true; blasSupport = true;
}).overrideAttrs }).overrideAttrs
(oldAttrs: rec { (oldAttrs: rec {
version = "8660"; version = "8683";
src = pkgs.fetchFromGitHub { src = pkgs.fetchFromGitHub {
owner = "ggml-org"; owner = "ggml-org";
repo = "llama.cpp"; repo = "llama.cpp";
tag = "b${version}"; tag = "b${version}";
# rev = "6ebf2e0d00d31acfc1a1fa9662e9a7d38bd07bf7"; # https://github.com/ggml-org/llama.cpp/pull/19970 # rev = "6ebf2e0d00d31acfc1a1fa9662e9a7d38bd07bf7"; # https://github.com/ggml-org/llama.cpp/pull/19970
hash = "sha256-4QwFMHpaYeD+aul/p5XFBRYivJ6WgI1kxecbrqFml6o="; hash = "sha256-cehplQ4utj8h3zpUqjsXmNiW2g+YylcNbvaRYVRP2/U=";
leaveDotGit = true; leaveDotGit = true;
postFetch = '' postFetch = ''
git -C "$out" rev-parse --short HEAD > $out/COMMIT git -C "$out" rev-parse --short HEAD > $out/COMMIT

5
machines/homepc/configuration.nix
View File

@@ -94,7 +94,7 @@
cmd: llama-server --port ''${PORT} -hf mudler/Qwen3.5-35B-A3B-APEX-GGUF:Mini --ctx-size 128000 --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 -np 1 --fit-target 256 --ubatch-size 1024 -fa on --slots --slot-save-path /home/laythe/llamapcache --jinja -kvu --no-mmproj --swa-checkpoints 32 --no-kv-offload cmd: llama-server --port ''${PORT} -hf mudler/Qwen3.5-35B-A3B-APEX-GGUF:Mini --ctx-size 128000 --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 -np 1 --fit-target 256 --ubatch-size 1024 -fa on --slots --slot-save-path /home/laythe/llamapcache --jinja -kvu --no-mmproj --swa-checkpoints 32 --no-kv-offload
ttl: 2400 ttl: 2400
"Gemma4-26B-A4B": "Gemma4-26B-A4B":
cmd: llama-server --port ''${PORT} -hf unsloth/gemma-4-26B-A4B-it-GGUF:UD-IQ4_XS --ctx-size 128000 --temp 1.0 --top-p 0.95 --top-k 64 --min-p 0.00 -np 1 --fit-target 256 --ubatch-size 1024 -fa on --slots --slot-save-path /home/laythe/llamapcache --jinja -kvu --no-mmproj --swa-checkpoints 32 --no-kv-offload cmd: llama-server --port ''${PORT} -hf mudler/gemma-4-26B-A4B-it-heretic-APEX-GGUF:Mini --ctx-size 128000 --temp 1.0 --top-p 0.95 --top-k 64 --min-p 0.00 -np 1 --fit-target 256 --ubatch-size 1024 -fa on --slots --slot-save-path /home/laythe/llamapcache --jinja -kvu --no-mmproj --swa-checkpoints 32 --no-kv-offload
ttl: 2400 ttl: 2400
"Qwen3-1.7B-GGUF": "Qwen3-1.7B-GGUF":
cmd: llama-server --port ''${PORT} -hf unsloth/Qwen3-1.7B-GGUF:Q4_K_S --ctx-size 32768 --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 -fa on --jinja -kvu cmd: llama-server --port ''${PORT} -hf unsloth/Qwen3-1.7B-GGUF:Q4_K_S --ctx-size 32768 --temp 0.6 --top-p 0.95 --top-k 20 --min-p 0.00 -fa on --jinja -kvu
@@ -120,6 +120,9 @@
"Qwen3.5-4B-Non-Thinking": "Qwen3.5-4B-Non-Thinking":
cmd: "llama-server --port ''${PORT} -hf unsloth/Qwen3.5-4B-GGUF:Q4_K_M --ctx-size 64000 --temp 0.7 --top-p 0.8 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256" cmd: "llama-server --port ''${PORT} -hf unsloth/Qwen3.5-4B-GGUF:Q4_K_M --ctx-size 64000 --temp 0.7 --top-p 0.8 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256"
ttl: 300 ttl: 300
"Gemma4-E4B":
cmd: "llama-server --port ''${PORT} -hf Abhiray/gemma-4-E4B-it-heretic-GGUF:Q4_K_M --ctx-size 64000 --temp 0.7 --top-p 0.8 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256"
ttl: 300
"Qwen3.5-0.8B-Non-Thinking": "Qwen3.5-0.8B-Non-Thinking":
cmd: "llama-server --port ''${PORT} -hf unsloth/Qwen3.5-0.8B-GGUF:Q4_K_M --ctx-size 64000 --temp 0.7 --top-p 0.8 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256" cmd: "llama-server --port ''${PORT} -hf unsloth/Qwen3.5-0.8B-GGUF:Q4_K_M --ctx-size 64000 --temp 0.7 --top-p 0.8 --top-k 20 --min-p 0.00 -fa on --jinja -kvu -np 1 --fit-target 256"
ttl: 300''; ttl: 300'';