* fix: Cookbook local GGUF serving inside Docker
Cookbook’s in-container GGUF serve flow had multiple Docker-specific breakages that made local llama.cpp models fail or register against the wrong endpoint.
Fixes included here:
use the scanned model cache root when generating GGUF serve commands instead of hardcoding $HOME/.cache/huggingface/hub
fix malformed llama.cpp preflight build lines that generated invalid bash in serve runner scripts
preserve loopback model URLs inside Docker when the target port is already reachable from the Odysseus container, instead of rewriting them unconditionally to host.docker.internal
Before this change, Docker local serves could fail in several ways:
Cookbook pointed llama.cpp at the wrong GGUF path
generated serve runner scripts crashed before launch with a shell syntax error
successfully started in-container model servers were auto-registered as host.docker.internal: instead of localhost/127.0.0.1
This makes the Docker Cookbook path work as expected for: downloaded GGUF -> local llama.cpp serve -> endpoint registration
* test: add test for docker-local endpoint rewrites
* Cookbook: Engine filter + intelligent hardware-computed serve profiles
Two related Cookbook serving improvements for accurate, hardware-aware model
serving (especially on consumer GPUs that can only run GGUF/llama.cpp).
Engine filter
- New "Engine" dropdown (All / llama.cpp / vLLM / SGLang) beside the quant
picker. Pure client-side view filter over the fetched list via the same
_detectBackend() the serve commands use, so what you filter to is exactly what
would launch. Re-renders from cache (no refetch). Empty-state message + the
instant-cache-paint path account for it too.
Intelligent serve profiles (Quality / Balanced / Speed)
- services/hwfit/profiles.py: compute_serve_profiles() turns detected VRAM +
model size into concrete llama.cpp flags (n_gpu_layers, n_cpu_moe, cache-type,
context). Encodes the by-hand tuning: a too-big MoE offloads experts to CPU
instead of failing; a model that fits stays fully on GPU; quant tracks profile
intent; vision models keep image-encoder headroom. Reuses models.py VRAM math
so filtering and serving agree on what fits. Pure/deterministic (no t/s claims
— partial-offload speed isn't reliably predictable; fit is what's computed).
- /api/hwfit/profiles endpoint returns the profiles + the model's trained
context limit, with loose name matching (strips org/ prefix, -GGUF suffix,
quant tag) so a local GGUF folder name resolves to its catalog entry.
- _buildServeCmd (llama.cpp) now emits --n-cpu-moe / --flash-attn /
--cache-type-k/v when set, with llama-cpp-python fallback equivalents. It
previously only set -ngl/-c, which is why it OOM'd or ran slow.
- Serve panel: profile chips that fill the fields on click, plus CPU-MoE / KV
Cache / Flash Attn fields. Context is clamped to the model's trained limit
(and an absolute 1M sanity ceiling) on type/blur/profile-load and at launch —
fixes a crash where a stale 256k/16M preset + quantized KV cache caused an
amdgpu ErrorDeviceLost.
Tests: tests/test_serve_profiles.py (7) — offload vs full-GPU fit, never exceed
VRAM, context cap, launchable flags, vision headroom, no-GPU empty.
Checks: py_compile + node --check pass; pytest test_serve_profiles + test_hwfit_amd
green; verified live on an RDNA4 box (gfx1200) — Balanced lands ~ncm18 q4 128k,
matching hand-tuning.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Cookbook: make column-header sorting discoverable (incl. Newest)
Sorting in Cookbook is via clickable column headers (pewds' design), but the
headers had no visual cue that they're interactive — so sorting in general, and
the Newest sort on the Model header specifically, was undiscoverable.
- Style sortable headers as interactive: pointer cursor, hover underline, and
the active sort column bolded/highlighted. There was no CSS for
.hwfit-sortable / .hwfit-sort-active at all; this helps every existing sort,
not just Newest.
- The Model column header sorts by release_date (newest first), reusing the
existing header-click sort wiring and the "newest" SORT_KEY.
No new sort control — uses the existing column-header paradigm.
Checks: node --check passes.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Cookbook serve profiles: keep the on-disk file's quant fixed (don't propose Q6/Q2)
In the Serve tab the model is a specific GGUF file already on disk, so its quant
can't change — but the profiles were suggesting "Quality · Q6_K" / "Speed · Q2_K"
as if you could re-quantize it. That's meaningless when serving a fixed file.
- compute_serve_profiles gains serve_weights_gb / serve_quant. When set (SERVE
mode), the quant is locked to the file's and profiles differ only in the real
serving knobs — n_cpu_moe, KV-cache type, context. _weights_gb / _cpu_moe_for_budget
use the file's actual size instead of a quant-derived estimate. DOWNLOAD mode
(no override) still varies the quant to show download options.
- /api/hwfit/profiles accepts serve_weights_gb & serve_quant.
- The Serve panel parses the file's size (from m.size "20.6 GB") and quant (from
the repo/file name) and passes them, so profiles match what's actually served.
Result for a 20.6 GB Q4_K_M file: all three profiles stay Q4_K_M and differ by
KV/ctx/offload (Quality q8 KV 128k ncm21, Balanced q4 128k ncm17, Speed q4 32k
ncm15) — no nonsensical quant changes.
Tests: test_serve_mode_keeps_fixed_quant. Full serve-profile suite green (9).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* Cookbook serve: Vision toggle (auto-find mmproj) + live VRAM/RAM-spillover monitor
Two serve-panel additions:
1. **Vision toggle.** A "Vision" checkbox that serves the model with its
multimodal projector so it can read images. The mmproj path is resolved at
runtime (find mmproj-*.gguf next to the model), so dropping an mmproj file in
the model folder makes the toggle just work; `--mmproj … --image-max-tokens
1024` (native) / `--clip_model_path` (llama-cpp-python) only when on + found.
2. **Live GPU-memory monitor.** A readout that polls /api/cookbook/gpus every 4s
while the panel is open and shows VRAM used/total/%, free, and — crucially on
a discrete card — **RAM spillover** (AMD gtt_used_mb), with a plain-language
health hint: green/healthy, amber/tight, red/"spilled to RAM — slow (raise
CPU MoE or lower context)". Surfaces gtt_used_mb from the gpus endpoint
(previously read for total only and discarded for 'used').
Lets you see at a glance whether a config fits VRAM (fast) or is paging to system
RAM over PCIe (slow) instead of guessing.
Checks: node --check + py_compile pass.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Two small polish items in the Cookbook Serve panel.
Saved-config badge
The little count badge next to the Save button ("3 ▾" etc.) had a
generic "Saved launch configs" tooltip, so the number reads like a
notification dot. Make it spell out what it is and what clicking does:
"3 saved launch configs for <model> — click ▾ to load or delete"
(and "No saved launch configs for <model> yet — click Save to add
one" when empty). Tooltip stays in sync via _updateSavedToggleLabel
so save/delete updates both the count and the hint.
GPU chip on mixed-GPU boxes (#711)
The chip label was `${gpuCount}x ${gpu_name}`, where gpu_name is
just gpus[0].name — so a 4090 + 3060 reads as "2x RTX 4090". The
backend already emits gpu_groups (identical cards grouped, used by
the serve flow to pin CUDA_VISIBLE_DEVICES) and a per-card gpus[]
array, so use them:
- Label renders each homogeneous pool: "1× RTX 4090 + 1× RTX 3060".
Homogeneous setups keep the existing "2× RTX 4090" form.
- Tooltip lists each GPU with its index + VRAM, useful for picking
the right device when launching.
Refs #711.
The global Escape arbiter in ui.js only sees `.modal` elements, so the many
ad-hoc dropdowns and context popups that are built on the fly and appended to
<body> ignored Escape entirely: document-library card/chat menus, chat
context/stats/overflow popups, cookbook serve & running menus, calendar event
menus, and compare pane menus.
Add a small DOM-free dismissal registry (static/js/escMenuStack.js). Menus
register a dismiss callback while open, and the arbiter closes the
most-recently-opened one first, so a menu opened over a modal closes before the
modal. bindMenuDismiss() wires the ubiquitous "append-to-body, close on outside
click" idiom to both the outside-click listener and the Escape stack in one
call, and dismissOrRemove() lets the pre-existing bulk removers (scroll/swipe/
modal-dismiss cleanup, reopen sweeps) tear a menu down through its real teardown
instead of orphaning its stack entry.
Covers ~14 menus across documentLibrary, chatRenderer, cookbookServe,
cookbookRunning, calendar, and compare/panes. Every teardown path — item click,
outside click, swipe, toggle, rebuild, bulk cleanup — routes through the
registry so no entry is ever stranded.
tests/test_esc_menu_stack_js.py pins the registry's LIFO and
exactly-one-per-press guarantees (node-driven; skips when node is absent).
Two bugs hid the popup that opens on double-click (or right-click) of
a GPU button in the Serve panel:
1. z-index 240 vs the cookbook modal at 260 — popup rendered behind
the modal it was spawned from.
2. Horizontal position was just `button.left`, with no clamp against
the viewport. GPU buttons sit near the right edge of the modal, so
the popup got anchored at a left that pushed most of its body past
the viewport's right edge.
Switch the popup to position:fixed (escapes scrolling / transform
stacking contexts on any ancestor), bump z-index to 10010 (above the
themed-confirm / overlay layer that sits around 9000-10000), and
clamp left/top after measuring the rendered size — including flipping
above the button if there isn't room below. The popup is now fully
visible regardless of which GPU button it's anchored to or how
narrow the viewport is.
* Add Apple Silicon (Metal) GPU detection and unified-memory fit tuning
hardware.py detects Apple Silicon locally and over SSH, reporting
backend=metal, the chip name, and a RAM-scaled fraction of unified
memory as the usable GPU budget. fit.py gains an M1-M4 memory-bandwidth
table for realistic tok/s and drops vLLM-only formats (AWQ/GPTQ/FP8)
that can't be served on Metal.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
(cherry picked from commit 32ac81dbc680361463a088dae867d555d5a79c3b)
* Generate macOS/Metal serve commands and surface the Metal GPU
cookbook_routes.py adds a macOS serve path (Ollama, Metal-aware
llama.cpp build using `sysctl hw.ncpu` instead of `nproc`, and a clear
error if vLLM is attempted). The frontend defaults Metal serving to
llama.cpp and offers llama.cpp/Ollama instead of vLLM/SGLang. The
odysseus-cookbook CLI's `gpus` command reports the Metal GPU via
sysctl/vm_stat.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
(cherry picked from commit 4ba01ce25d256ae032029898f361c824a34fcd4b)
* Add launchd LaunchAgent for macOS (systemd equivalent)
com.odysseus.ui.plist + install-service-macos.sh run Odysseus at login
and restart on crash, the macOS counterpart to odysseus-ui.service. The
installer auto-fills paths from the venv, so there's no hand-editing.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
(cherry picked from commit 3d4b6b2c7b8b31af32201ed278115df9a559dea9)
* Document macOS install (brew, Ollama, AirPlay port, launchd)
README + setup.py cover the Homebrew / Apple Silicon path: brew install
python@3.11 tmux ollama, Metal serving via Ollama/llama.cpp, the launchd
service, and the macOS AirPlay Receiver conflict on ports 7000/5000.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
(cherry picked from commit 8dc9a3578a1726f070ed9f75c0958ae291a6d966)
* Add downloadable macOS launcher app builder
build-macos-app.sh generates dist/Odysseus.app and a drag-to-Applications
dist/Odysseus.dmg. The app starts the local server from this repo's venv and
opens the UI in a chrome-less app window (Chromium --app mode, falling back to
the default browser). It's a launcher wrapper — it drives the venv rather than
bundling Python — so the install path is baked in at build time.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
(cherry picked from commit 7927940c3810ee34640803b198d334a6ac93474d)
* Harden macOS Cookbook support: hide MLX, fix Metal build cache
Builds on the adopted PR #213 macOS/Metal work with two fixes and tests:
- fit.py: always drop MLX-quantized models. Odysseus only generates serve
commands for llama.cpp/Ollama (Metal) and vLLM/SGLang (CUDA); MLX needs the
mlx_lm runtime and the catalog's MLX repos ship no GGUF alternative, so they
were surfaced on Apple Silicon but could never be served.
- cookbook_routes.py (macOS branch only): `rm -rf build` before configure so a
poisoned CMakeCache from a prior failed CUDA attempt can't make every later
build fail; explicit -DCMAKE_BUILD_TYPE=Release; a clear "brew install cmake"
hint if cmake is missing. Linux/CUDA path unchanged.
- tests/test_hwfit_macos.py: MLX hidden on metal, MLX still hidden on CUDA
(regression guard), Metal detection on Apple Silicon, and skipped on
Linux/Intel (proves non-macOS detection is untouched).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* Propagate unified_memory flag and document macOS GPU/Docker caveat
- hardware.py: detect_system now carries the unified_memory flag from GPU
detection into the system dict (it was set by _detect_apple_silicon / AMD-APU
detection but dropped during result assembly, so the API always reported
null). Lets callers distinguish unified from discrete VRAM.
- README: prominent warning that Docker on Apple Silicon can't reach the Metal
GPU (runs a Linux VM) — Cookbook must run natively for GPU serving; fix stale
text that said Cookbook recommends MLX models (now hidden as unservable).
- test: detect_system propagates unified_memory.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* Put Odysseus's venv bin on PATH for cookbook runners
Native (non-Docker) installs run from a virtualenv whose bin holds the `hf` CLI
and `python3` the cookbook download/serve tmux scripts shell out to. Those
scripts start in a fresh login shell with the venv NOT activated, so on a native
macOS install `hf download` failed with "hf: command not found" — and the
`pip --user` self-heal missed because macOS has no bare `pip` command.
- cookbook_helpers.py: _local_tooling_path_export() — pure helper returning a
PATH export for the running interpreter's bin dir (escaped for double quotes).
- cookbook_routes.py: download + serve runners prepend that dir on local runs
(gated off SSH/Windows); swap the `pip` install fallbacks to `python3 -m pip`.
- tests: helper output for normal and spaced paths.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* Document macOS llama.cpp serving prerequisites
Clarify the two serving paths on Apple Silicon: the recommended zero-build
route (brew install llama.cpp ships a Metal llama-server Cookbook finds on PATH),
and the from-source fallback, which requires cmake + Xcode Command Line Tools.
Without those the build is skipped and serving silently degrades to a slow CPU
build, so new users now know to install them (or use the prebuilt) up front.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* Recommend only GGUF-servable models on Metal
Apple Silicon's only serving engines are llama.cpp and Ollama, both GGUF-only
(vLLM/SGLang are CUDA/ROCm and don't run on macOS). The catalog tags raw
safetensors repos with a default Q4_K_M quant, so the fit-ranking was
recommending ~397/501 models that have no GGUF and fail to serve on Metal with
"No GGUF found" (e.g. microsoft/Phi-mini-MoE-instruct).
Drop any model without a real GGUF (is_gguf/gguf_sources) on Apple Silicon —
subsumes the previous AWQ/GPTQ/FP8 special-case into one rule. On CUDA these
stay visible since vLLM serves safetensors directly. Metal recommendations go
501 -> 104, all actually servable.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* Remove macOS launchd LaunchAgent (cherry-picked extra)
Drop the launchd service from the PR #213 cherry-picks: the
install-service-macos.sh installer, the com.odysseus.ui.plist template, and the
README section documenting them. Tangential to the core Cookbook/Metal support
and not wanted. The build-macos-app.sh launcher is kept.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* Add one-command macOS quick start (start-macos.sh)
Running Odysseus natively on a Mac previously meant ~7 manual terminal steps
(brew deps, venv, activate, pip, setup.py, uvicorn with the right port) — not
friendly for a generic macOS user, and the native run is required because Docker
on macOS can't reach the Metal GPU.
- start-macos.sh: installs Homebrew deps (python@3.11, tmux, prebuilt Metal
llama.cpp), creates the venv, installs requirements, runs setup, and launches
on a non-AirPlay port (7860). Idempotent; re-run to start again.
- README: the Apple Silicon section now leads with this one-command quick start
and the clickable .app, with engine/port/manual details folded into a
collapsible block. Added a pointer at the top of the manual-install section.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* macOS quick start: auto-open browser when ready
The "open this URL" line scrolled out of view as uvicorn kept logging after it,
so users missed it. Now start-macos.sh waits (in the background) until the
server accepts connections, prints a boxed "ready" banner at that point (i.e.
after the startup burst, not before), and opens the URL in the default browser
automatically. Skippable with ODYSSEUS_NO_OPEN=1 for headless/SSH use.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* Don't assume/force a specific Python version on macOS
The README claimed "system Python is 3.9" — a machine-specific generalization
that's often wrong (macOS ships no recent Python by default; many users already
have 3.11+). Make it generic, and make start-macos.sh detect an existing
Python 3.11+ and use it, only installing python@3.11 when none is found instead
of forcing it on top of the user's Python.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* Align start-macos.sh venv path with build-macos-app.sh
start-macos.sh created the environment in .venv/, but build-macos-app.sh and
the manual install steps use venv/ — so the clickable .app wouldn't reuse the
quick-start's environment and would rebuild a second one. Use venv/ everywhere.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* README: state clearly that MLX is unsupported on Apple Silicon
Odysseus has no mlx_lm runtime; it serves GGUF (llama.cpp/Ollama) and CUDA
(vLLM/SGLang) only. MLX-only models can't run on a Mac and are hidden from
Cookbook — make that explicit in both the quick start and the details.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* start-macos.sh: build the venv with an arm64 Python on Apple Silicon
A clean-room run surfaced this: with a universal2/x86 Python (e.g. the
python.org installer under /usr/local), the venv's compiled extensions install
as arm64 but get loaded as x86_64 when launched from the .app bundle, so it
crashes with "incompatible architecture (have arm64, need x86_64)". The terminal
run happened to work only because a universal binary defaults to arm64 there.
On Apple Silicon, look only under /opt/homebrew (arm64-only) for the build
Python, and install Homebrew's python@3.11 if none is present — so the venv is
arm64-only and launches correctly from both the terminal and the .app. Intel
and non-mac paths are unchanged. Verified end-to-end in a clean clone: .app now
boots on Metal with no arch error.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
* Address dev-exp review: macOS setup robustness + doc/UX fixes
From the voltagent dev-exp review of the branch:
- README: fix broken anchor links (the em-dash heading produced a slug the links
didn't match); simplify the heading to a stable slug.
- cookbook_routes.py: add /opt/homebrew/bin and /usr/local/bin to the serve PATH
so a brew-installed llama-server/ollama is found instead of falling back to a
slow source build.
- start-macos.sh: guard against an empty Python path; fail fast with a clear
message on port-in-use; ERR trap with a "safe to re-run" message; show pip
progress (drop --quiet on the slow requirements install); stop the background
browser-opener cleanly on exit/Ctrl+C (no orphaned poller).
- setup.py: bind hint to 127.0.0.1; suppress the manual run-hint when launched
by start-macos.sh (ODYSSEUS_SKIP_RUN_HINT) so the URL isn't contradictory.
- build-macos-app.sh: the .app only opens the browser once the server is
actually ready (not after the readiness timeout).
- cookbookServe.js: drop "Diffusers" from the Metal backend picker —
diffusion_server.py is CUDA-only, so it was an unservable option on macOS.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
---------
Co-authored-by: yunggilja <yunggilja@gmail.com>
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
