95efb8a562
* feat(backend): add turboquant llama.cpp-fork backend
turboquant is a llama.cpp fork (TheTom/llama-cpp-turboquant, branch
feature/turboquant-kv-cache) that adds a TurboQuant KV-cache scheme.
It ships as a first-class backend reusing backend/cpp/llama-cpp sources
via a thin wrapper Makefile: each variant target copies ../llama-cpp
into a sibling build dir and invokes llama-cpp's build-llama-cpp-grpc-server
with LLAMA_REPO/LLAMA_VERSION overridden to point at the fork. No
duplication of grpc-server.cpp — upstream fixes flow through automatically.
Wires up the full matrix (CPU, CUDA 12/13, L4T, L4T-CUDA13, ROCm, SYCL
f32/f16, Vulkan) in backend.yml and the gallery entries in index.yaml,
adds a tests-turboquant-grpc e2e job driven by BACKEND_TEST_CACHE_TYPE_K/V=q8_0
to exercise the KV-cache config path (backend_test.go gains dedicated env
vars wired into ModelOptions.CacheTypeKey/Value — a generic improvement
usable by any llama.cpp-family backend), and registers a nightly auto-bump
PR in bump_deps.yaml tracking feature/turboquant-kv-cache.
scripts/changed-backends.js gets a special-case so edits to
backend/cpp/llama-cpp/ also retrigger the turboquant CI pipeline, since
the wrapper reuses those sources.
* feat(turboquant): carry upstream patches against fork API drift
turboquant branched from llama.cpp before upstream commit 66060008
("server: respect the ignore eos flag", #21203) which added the
`logit_bias_eog` field to `server_context_meta` and a matching
parameter to `server_task::params_from_json_cmpl`. The shared
backend/cpp/llama-cpp/grpc-server.cpp depends on that field, so
building it against the fork unmodified fails.
Cherry-pick that commit as a patch file under
backend/cpp/turboquant/patches/ and apply it to the cloned fork
sources via a new apply-patches.sh hook called from the wrapper
Makefile. Simplifies the build flow too: instead of hopping through
llama-cpp's build-llama-cpp-grpc-server indirection, the wrapper now
drives the copied Makefile directly (clone -> patch -> build).
Drop the corresponding patch whenever the fork catches up with
upstream — the build fails fast if a patch stops applying, which
is the signal to retire it.
* docs: add turboquant backend section + clarify cache_type_k/v
Document the new turboquant (llama.cpp fork with TurboQuant KV-cache)
backend alongside the existing llama-cpp / ik-llama-cpp sections in
features/text-generation.md: when to pick it, how to install it from
the gallery, and a YAML example showing backend: turboquant together
with cache_type_k / cache_type_v.
Also expand the cache_type_k / cache_type_v table rows in
advanced/model-configuration.md to spell out the accepted llama.cpp
quantization values and note that these fields apply to all
llama.cpp-family backends, not just vLLM.
* feat(turboquant): patch ggml-rpc GGML_OP_COUNT assertion
The fork adds new GGML ops bringing GGML_OP_COUNT to 97, but
ggml/include/ggml-rpc.h static-asserts it equals 96, breaking
the GGML_RPC=ON build paths (turboquant-grpc / turboquant-rpc-server).
Carry a one-line patch that updates the expected count so the
assertion holds. Drop this patch whenever the fork fixes it upstream.
* feat(turboquant): allow turbo* KV-cache types and exercise them in e2e
The shared backend/cpp/llama-cpp/grpc-server.cpp carries its own
allow-list of accepted KV-cache types (kv_cache_types[]) and rejects
anything outside it before the value reaches llama.cpp's parser. That
list only contains the standard llama.cpp types — turbo2/turbo3/turbo4
would throw "Unsupported cache type" at LoadModel time, meaning
nothing the LocalAI gRPC layer accepted was actually fork-specific.
Add a build-time augmentation step (patch-grpc-server.sh, called from
the turboquant wrapper Makefile) that inserts GGML_TYPE_TURBO2_0/3_0/4_0
into the allow-list of the *copied* grpc-server.cpp under
turboquant-<flavor>-build/. The original file under backend/cpp/llama-cpp/
is never touched, so the stock llama-cpp build keeps compiling against
vanilla upstream which has no notion of those enum values.
Switch test-extra-backend-turboquant to set
BACKEND_TEST_CACHE_TYPE_K=turbo3 / _V=turbo3 so the e2e gRPC suite
actually runs the fork's TurboQuant KV-cache code paths (turbo3 also
auto-enables flash_attention in the fork). Picking q8_0 here would
only re-test the standard llama.cpp path that the upstream llama-cpp
backend already covers.
Refresh the docs (text-generation.md + model-configuration.md) to
list turbo2/turbo3/turbo4 explicitly and call out that you only get
the TurboQuant code path with this backend + a turbo* cache type.
* fix(turboquant): rewrite patch-grpc-server.sh in awk, not python3
The builder image (ubuntu:24.04 stage-2 in Dockerfile.turboquant)
does not install python3, so the python-based augmentation step
errored with `python3: command not found` at make time. Switch to
awk, which ships in coreutils and is already available everywhere
the rest of the wrapper Makefile runs.
* Apply suggestion from @mudler
Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com>
---------
Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com>
LocalAI website
LocalAI documentation website
Requirement
In this project, the Docsy theme component is pulled in as a Hugo module, together with other module dependencies:
$ hugo mod graph
hugo: collected modules in 566 ms
hugo: collected modules in 578 ms
github.com/google/docsy-example github.com/google/docsy@v0.5.1-0.20221017155306-99eacb09ffb0
github.com/google/docsy-example github.com/google/docsy/dependencies@v0.5.1-0.20221014161617-be5da07ecff1
github.com/google/docsy/dependencies@v0.5.1-0.20221014161617-be5da07ecff1 github.com/twbs/bootstrap@v4.6.2+incompatible
github.com/google/docsy/dependencies@v0.5.1-0.20221014161617-be5da07ecff1 github.com/FortAwesome/Font-Awesome@v0.0.0-20220831210243-d3a7818c253f
If you want to do SCSS edits and want to publish these, you need to install PostCSS
npm install
Running the website locally
Building and running the site locally requires a recent extended version of Hugo.
You can find out more about how to install Hugo for your environment in our
Getting started guide.
Once you've made your working copy of the site repo, from the repo root folder, run:
hugo server
Running a container locally
You can run docsy-example inside a Docker
container, the container runs with a volume bound to the docsy-example
folder. This approach doesn't require you to install any dependencies other
than Docker Desktop on
Windows and Mac, and Docker Compose
on Linux.
-
Build the docker image
docker-compose build -
Run the built image
docker-compose upNOTE: You can run both commands at once with
docker-compose up --build. -
Verify that the service is working.
Open your web browser and type
http://localhost:1313in your navigation bar, This opens a local instance of the docsy-example homepage. You can now make changes to the docsy example and those changes will immediately show up in your browser after you save.
Cleanup
To stop Docker Compose, on your terminal window, press Ctrl + C.
To remove the produced images run:
docker-compose rm
For more information see the Docker Compose documentation.
Troubleshooting
As you run the website locally, you may run into the following error:
➜ hugo server
INFO 2021/01/21 21:07:55 Using config file:
Building sites … INFO 2021/01/21 21:07:55 syncing static files to /
Built in 288 ms
Error: Error building site: TOCSS: failed to transform "scss/main.scss" (text/x-scss): resource "scss/scss/main.scss_9fadf33d895a46083cdd64396b57ef68" not found in file cache
This error occurs if you have not installed the extended version of Hugo. See this section of the user guide for instructions on how to install Hugo.
Or you may encounter the following error:
➜ hugo server
Error: failed to download modules: binary with name "go" not found
This error occurs if you have not installed the go programming language on your system.
See this section of the user guide for instructions on how to install go.