* feat(llama.cpp): Enable decentralized, distributed inference
As https://github.com/mudler/LocalAI/pull/2324 introduced distributed inferencing thanks to
@rgerganov implementation in https://github.com/ggerganov/llama.cpp/pull/6829 in upstream llama.cpp, now
it is possible to distribute the workload to remote llama.cpp gRPC server.
This changeset now uses mudler/edgevpn to establish a secure, distributed network between the nodes using a shared token.
The token is generated automatically when starting the server with the `--p2p` flag, and can be used by starting the workers
with `local-ai worker p2p-llama-cpp-rpc` by passing the token via environment variable (TOKEN) or with args (--token).
As per how mudler/edgevpn works, a network is established between the server and the workers with dht and mdns discovery protocols,
the llama.cpp rpc server is automatically started and exposed to the underlying p2p network so the API server can connect on.
When the HTTP server is started, it will discover the workers in the network and automatically create the port-forwards to the service locally.
Then llama.cpp is configured to use the services.
This feature is behind the "p2p" GO_FLAGS
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* go mod tidy
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* ci: add p2p tag
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* better message
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
---------
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
feat(functions): support mixed JSON BNF grammar
This PR provides new options to control how functions are extracted from
the LLM, and also provides more control on how JSON grammars can be used
(also in conjunction).
New YAML settings introduced:
- `grammar_message`: when enabled, the generated grammar can also decide
to push strings and not only JSON objects. This allows the LLM to pick
to either respond freely or using JSON.
- `grammar_prefix`: Allows to prefix a string to the JSON grammar
definition.
- `replace_results`: Is a map that allows to replace strings in the LLM
result.
As an example, consider the following settings for Hermes-2-Pro-Mistral,
which allow extracting both JSON results coming from the model, and the
ones coming from the grammar:
```yaml
function:
# disable injecting the "answer" tool
disable_no_action: true
# This allows the grammar to also return messages
grammar_message: true
# Suffix to add to the grammar
grammar_prefix: '<tool_call>\n'
return_name_in_function_response: true
# Without grammar uncomment the lines below
# Warning: this is relying only on the capability of the
# LLM model to generate the correct function call.
# no_grammar: true
# json_regex_match: "(?s)<tool_call>(.*?)</tool_call>"
replace_results:
"<tool_call>": ""
"\'": "\""
```
Note: To disable entirely grammars usage in the example above, uncomment the
`no_grammar` and `json_regex_match`.
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
When enabling grammar with functions, it might be useful to
allow more flexibility to support models that are fine-tuned against returning
function calls of the form of { "name": "function_name", "arguments" {...} }
rather then { "function": "function_name", "arguments": {..} }.
This might call out to a more generic approach later on, but for the moment being we can easily support both
as we have just to specific different types.
If needed we can expand on this later on
Signed-off-by: mudler <mudler@localai.io>
* feat(ui): allow to set system prompt for chat
Make also the models in the index clickable, and display as table
Fixes#2257
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* feat(vision): support also png with base64 input
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* feat(ui): support vision and upload of files
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* display the processed image
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* make trust remote code stand out
Signed-off-by: mudler <mudler@localai.io>
* feat(ui): track in progress job across index/model gallery
Signed-off-by: mudler <mudler@localai.io>
* minor fixups
Signed-off-by: mudler <mudler@localai.io>
---------
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Signed-off-by: mudler <mudler@localai.io>
* ux: change welcome when there are no models installed
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* ux: filter
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* ux: show tags in filter
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* wip
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* make tags clickable
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* allow to delete models from the list
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* ui: display icon of installed models
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* gallery: remove gallery file when removing model
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* feat(gallery): show a re-install button
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* make filter buttons, rename Gallery field
Signed-off-by: mudler <mudler@localai.io>
* show again buttons at end of operations
Signed-off-by: mudler <mudler@localai.io>
---------
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Signed-off-by: mudler <mudler@localai.io>
* feat(initializer): do not specify backends to autoload
We can simply try to autoload the backends extracted in the asset dir.
This will allow to build variants of the same backend (for e.g. with different instructions sets),
so to have a single binary for all the variants.
Signed-off-by: mudler <mudler@localai.io>
* refactor(prepare): refactor out llama.cpp prepare steps
Make it so are idempotent and that we can re-build
Signed-off-by: mudler <mudler@localai.io>
* [TEST] feat(build): build noavx version along
Signed-off-by: mudler <mudler@localai.io>
* build: make build parallel
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* build: do not override CMAKE_ARGS
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* build: add fallback variant
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* Fixups
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* fix(huggingface-langchain): fail if no token is set
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* fix(huggingface-langchain): rename
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* fix: do not autoload local-store
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* fix: give priority between the listed backends
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
---------
Signed-off-by: mudler <mudler@localai.io>
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* start breaking up the giant channel refactor now that it's better understood - easier to merge bites
Signed-off-by: Dave Lee <dave@gray101.com>
* add concurrency and base64 back in, along with new base64 tests.
Signed-off-by: Dave Lee <dave@gray101.com>
* Automatic rename of whisper.go's Result to TranscriptResult
Signed-off-by: Dave Lee <dave@gray101.com>
* remove pkg/concurrency - significant changes coming in split 2
Signed-off-by: Dave Lee <dave@gray101.com>
* fix comments
Signed-off-by: Dave Lee <dave@gray101.com>
* add list_model service as another low-risk service to get it out of the way
Signed-off-by: Dave Lee <dave@gray101.com>
* split backend config loader into seperate file from the actual config struct. No changes yet, just reduce cognative load with smaller files of logical blocks
Signed-off-by: Dave Lee <dave@gray101.com>
* rename state.go ==> application.go
Signed-off-by: Dave Lee <dave@gray101.com>
* fix lost import?
Signed-off-by: Dave Lee <dave@gray101.com>
---------
Signed-off-by: Dave Lee <dave@gray101.com>
* feat(gallery): op now supports deletion of models
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* Wire things with WebUI(WIP)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* minor improvements
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
---------
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* fix: update dockerignore and gitignore to exclude sonarqube work dir
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* fix: remove useless equality check
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* fix: use sonarqube Dockerfile recommendations
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
---------
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* feat: enable polling configs for systems with broken fsnotify (docker volumes on windows)
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* fix: update logging to make it clear that the config file is being polled
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
---------
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* feat: refactor the dynamic json configs for api_keys and external_backends
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* fix: remove commented code
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
---------
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com>
Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com>
Signed-off-by: Dave <dave@gray101.com>
---------
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Signed-off-by: Dave <dave@gray101.com>
Co-authored-by: Dave <dave@gray101.com>
* feat: migrate to alecthomas/kong for CLI
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* feat: bring in new flag for granular log levels
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* chore: go mod tidy
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* feat: allow loading cli flag values from ["./localai.yaml", "~/.config/localai.yaml", "/etc/localai.yaml"] in that order
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* feat: load from .env file instead of a yaml file
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* feat: better loading for environment files
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* feat(doc): add initial documentation about configuration
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* fix: remove test log lines
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* feat: integrate new documentation into existing pages
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* feat: add documentation on .env files
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* fix: cleanup some documentation table errors
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* feat: refactor CLI logic out to it's own package under core/cli
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
---------
Signed-off-by: Chris Jowett <421501+cryptk@users.noreply.github.com>
* fix(seed): generate random seed per-request if -1 is set
Also update ci with new workflows and allow the aio tests to run with an
api key
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* docs(openvino): Add OpenVINO example
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
---------
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
* fix(grammar): Fix JSON mode and custom grammar
* tests(aio): add jsonmode test
* tests(aio): add functioncall test
* fix(aio): use hermes-2-pro-mistral as llm for CPU profile
* add phi-2-orange
* Initial implementation of assistants api
* Move load/save configs to utils
* Save assistant and assistantfiles config to disk.
* Add tsets for assistant api
* Fix models path spelling mistake.
* Remove personal go.mod information
---------
Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com>
* test with gguf instead of ggml. Updates testPrompt to match? Adds debugging line to Dockerfile that I've found helpful recently.
* fix testPrompt slightly
* Sad Experiment: Test GH runner without metal?
* break apart CGO_LDFLAGS
* switch runner
* upstream llama.cpp disables Metal on Github CI!
* missed a dir from clean-tests
* CGO_LDFLAGS
* tmate failure + NO_ACCELERATE
* whisper.cpp has a metal fix
* do the exact opposite of the name of this branch, but keep it around for unrelated fixes?
* add back newlines
* add tmate to linux for testing
* update fixtures
* timeout for tmate
Certain engines requires to know during model loading
if the embedding feature has to be enabled, however, it is impractical
to have to set it to ALL the backends that supports embeddings.
There are transformers and sentencentransformers that seamelessly handle
both cases, without having this settings to be explicitly enabled.
The case sussist only for ggml-based models that needs to enable
featuresets during model loading (and thus settings `embedding` is
required), however most of the other engines does not require this.
This change disables the check done at code side, making easier to use
embeddings by not having to specify explicitly `embeddings: true`.
Part of: https://github.com/mudler/LocalAI/issues/1373
* feat(elevenlabs): map elevenlabs API support to TTS
This allows elevenlabs Clients to work automatically with LocalAI by
supporting the elevenlabs API.
The elevenlabs server endpoint is implemented such as it is wired to the
TTS endpoints.
Fixes: https://github.com/mudler/LocalAI/issues/1809
* feat(openai/tts): compat layer with openai tts
Fixes: #1276
* fix: adapt tts CLI
* fix(defaults): set better defaults for inferencing
This changeset aim to have better defaults and to properly detect when
no inference settings are provided with the model.
If not specified, we defaults to mirostat sampling, and offload all the
GPU layers (if a GPU is detected).
Related to https://github.com/mudler/LocalAI/issues/1373 and https://github.com/mudler/LocalAI/issues/1723
* Adapt tests
* Also pre-initialize default seed
* feat(intel): add diffusers support
* try to consume upstream container image
* Debug
* Manually install deps
* Map transformers/hf cache dir to modelpath if not specified
* fix(compel): update initialization, pass by all gRPC options
* fix: add dependencies, implement transformers for xpu
* base it from the oneapi image
* Add pillow
* set threads if specified when launching the API
* Skip conda install if intel
* defaults to non-intel
* ci: add to pipelines
* prepare compel only if enabled
* Skip conda install if intel
* fix cleanup
* Disable compel by default
* Install torch 2.1.0 with Intel
* Skip conda on some setups
* Detect python
* Quiet output
* Do not override system python with conda
* Prefer python3
* Fixups
* exllama2: do not install without conda (overrides pytorch version)
* exllama/exllama2: do not install if not using cuda
* Add missing dataset dependency
* Small fixups, symlink to python, add requirements
* Add neural_speed to the deps
* correctly handle model offloading
* fix: device_map == xpu
* go back at calling python, fixed at dockerfile level
* Exllama2 restricted to only nvidia gpus
* Tokenizer to xpu
* core 1
* api/openai/files fix
* core 2 - core/config
* move over core api.go and tests to the start of core/http
* move over localai specific endpoints to core/http, begin the service/endpoint split there
* refactor big chunk on the plane
* refactor chunk 2 on plane, next step: port and modify changes to request.go
* easy fixes for request.go, major changes not done yet
* lintfix
* json tag lintfix?
* gitignore and .keep files
* strange fix attempt: rename the config dir?
This PR specifically introduces a `core` folder and moves the following packages over, without any other changes:
- `api/backend`
- `api/config`
- `api/options`
- `api/schema`
Once this is merged and we confirm there's no regressions, I can migrate over the remaining changes piece by piece to split up application startup, backend services, http, and mqtt as was the goal of the earlier PRs!
