MrSphay/odysseus
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Fix native Cookbook quant classification

Browse Source
This commit is contained in:
spooky
2026-06-02 14:07:20 +10:00
committed by GitHub
parent 65b5d65059
commit cd4f496cb4
6 changed files with 201 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
static/js/cookbook.js
View File

@@ -262,10 +262,10 @@ export function _detectBackend(model) {
const isRocm = sysBackend === 'rocm';
const isAppleSilicon = ['metal', 'mps', 'apple'].includes(sysBackend);
const _nm = `${model.repo_id || ''} ${model.path || ''} ${model.name || ''}`.toLowerCase();
if (!isAppleSilicon && (/\bmlx\b|mlx-|_mlx/i.test(_nm) || q.startsWith('MLX'))) {
if (/\bmlx\b|mlx-|_mlx/i.test(_nm) || q.startsWith('MLX')) {
return { backend: 'unsupported', label: 'Unsupported' };
}
const isAwqLike = /^AWQ|^GPTQ|^NVFP4/.test(q) || q === 'FP8' || /\b(awq|gptq|fp8|nvfp4)\b/i.test(_nm);
const isAwqLike = /^AWQ|^GPTQ|^NVFP4/.test(q) || ['FP8', 'FP4', 'MXFP4', 'NF4', 'INT4', 'INT8', 'W4A16', 'W8A8', 'W8A16'].includes(q) || /\b(awq|gptq|fp8|fp4|nvfp4|mxfp4|nf4|int4|int8|w4a16|w8a8|w8a16)\b/i.test(_nm);
const isGgufLike = model.is_gguf || /^Q[2-8]/.test(q) || /^IQ/.test(q) || q === 'GGUF' || _nm.includes('gguf');
// Image gen models → diffusers
@@ -291,7 +291,7 @@ export function _detectBackend(model) {
}
// Apple Silicon (Metal) → llama.cpp (GGUF). vLLM/SGLang are CUDA/ROCm-only and
// don't run on macOS; AWQ/GPTQ/FP8 (vLLM-only) models are already filtered out
// don't run on macOS; vLLM-native quantized models are already filtered out
// of metal Cookbook results, so llama.cpp is always the right engine here.
if (['metal', 'mps', 'apple'].includes(sysBackend)) {
return { backend: 'llamacpp', label: 'llama.cpp' };
@@ -1516,7 +1516,7 @@ function _renderRecipes() {
html += '<option value="Q4_K_M">Q4</option><option value="Q8_0">Q8</option>';
html += '<option value="Q6_K">Q6</option><option value="Q5_K_M">Q5</option>';
html += '<option value="Q3_K_M">Q3</option><option value="Q2_K">Q2</option>';
html += '<option value="AWQ-4bit">AWQ</option><option value="FP8">FP8</option>';
html += '<option value="AWQ-4bit">AWQ</option><option value="FP8">FP8</option><option value="FP4">FP4</option>';
html += '<option value="">Native</option></select>';
// Engine filter: show only models whose serve engine matches. "llama.cpp"
// (GGUF) runs everywhere incl. consumer AMD/Apple; vLLM/SGLang are CUDA /