turbollm 1.10.6 → 1.10.8

package/README.md

@@ -427,10 +427,14 @@ turbollm
 1. Detects your GPU and **downloads a matching `llama-server` build** (CUDA for NVIDIA, ROCm
    for AMD, Metal for Apple, SYCL for Intel, Vulkan otherwise — with a CPU fallback).
 2. Starts on <http://127.0.0.1:6996> and opens your browser.
-3. Drops you on the **Chat** screen, ready to load a model.
+3. Walks you through a short **setup wizard**: pick how you'll use it (casual chat, coding,
+   tinkering, or pro), get a model recommended for your actual hardware (or bring your own),
+   watch it download and load for real, and try it — before landing in **Chat** or **Code**.
+   Skippable from any step, and safe to close mid-download — reopen `/onboarding` any time
+   from **Models** to pick up where you left off.
 
-Then open **Models**, download or pick a GGUF, click **Load**, and start chatting. Stop the
-daemon any time with **Ctrl+C**.
+After that (or any time later), open **Models**, download or pick a GGUF, click **Load**, and
+start chatting. Stop the daemon any time with **Ctrl+C**.
 
 ---
 
@@ -464,8 +468,8 @@ switch between CUDA / ROCm / Metal / SYCL / Vulkan / CPU at any time (it downloa
 you choose, LM Studio-style).
 
 **Engine types.** **llama.cpp / GGUF**, **KoboldCpp** and **llamafile** (GGUF, every OS),
-**MLX** (macOS), and **vLLM** (Linux + NVIDIA) are all first-class engine kinds — install from
-the curated catalog, pick the right one per model, and switch from a single dropdown.
+**MLX** and **MLX-VLM** (macOS), and **vLLM** (Linux + NVIDIA) are all first-class engine kinds —
+install from the curated catalog, pick the right one per model, and switch from a single dropdown.
 
 **Fully supervised.** Every engine runs under a real state machine: health-gated readiness,
 graceful stop, an **idle auto-stop** watchdog, and **live logs + clear error surfacing** in

package/dist/{chunk-5MCP2ZAJ.js → chunk-H67DVIQT.js}

@@ -8,16 +8,18 @@ import { join } from "path";
 function engineAcceptsFormat(engineKind, format) {
   if (engineKind === "mlx") return format === "mlx";
   if (engineKind === "rapid-mlx") return format === "mlx";
+  if (engineKind === "mlx-vlm") return format === "mlx";
   if (engineKind === "vllm") return format === "mlx";
   return format === "gguf";
 }
 function engineRejectsAudioModel(engineKind) {
-  return engineKind === "rapid-mlx";
+  return engineKind === "rapid-mlx" || engineKind === "mlx-vlm";
 }
 var ENGINE_MODEL_ALIAS = "default_model";
 var RAPID_MLX_MODEL_ALIAS = "default";
-function engineModelAlias(engineKind) {
+function engineModelAlias(engineKind, modelPath) {
   if (engineKind === "rapid-mlx") return RAPID_MLX_MODEL_ALIAS;
+  if (engineKind === "mlx-vlm") return modelPath ?? null;
   return engineKind === "mlx" || engineKind === "vllm" || engineKind === "sglang" ? ENGINE_MODEL_ALIAS : null;
 }
 var VENDOR_DISPLAY = {
@@ -77,7 +79,7 @@ async function distill(d, material) {
   const ms = d.manager.status();
   const target = d.manager.target();
   if (ms.state !== "running" || !ms.model || !target) return { name: null, description: null, procedure: null };
-  const model = engineModelAlias(d.registry.active()?.kind ?? "") ?? ms.model.key;
+  const model = engineModelAlias(d.registry.active()?.kind ?? "", d.manager.currentOpts()?.modelPath) ?? ms.model.key;
   try {
     const res = await fetch(`${target}/v1/chat/completions`, {
       method: "POST",