Commit Graph

20 Commits

Author SHA1 Message Date
Ettore Di Giacinto
ea330d452d
models(gallery): add mistral-0.3 and command-r, update functions (#2388)
* models(gallery): add mistral-0.3 and command-r, update functions

Add also disable_parallel_new_lines to disable newlines in the JSON
output when forcing parallel tools. Some models (like mistral) might be
very sensible to that when being used for function calling.

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* models(gallery): add aya-23-8b

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

---------

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
2024-05-23 19:16:08 +02:00
Ettore Di Giacinto
491e1d752b
feat(functions): relax mixedgrammars (#2365)
* feat(functions): relax mixedgrammars

Extend even more the functionalities and when mixed mode is enabled,
tolerate also both strings and JSON in the result - in this case we make
sure that the JSON can be correctly parsed.

This also updates the examples and the gallery model to configure the
grammar.

The changeset also breaks current function/grammar configuration as it
reserves now a stanza in the YAML config.

For example:

```yaml
function:
  grammar:
    # This allows the grammar to also return messages
    mixed_mode: true
    # Suffix to add to the grammar
    # prefix: '<tool_call>\n'
    # Force parallel calls in the grammar
    # parallel_calls: true
```

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* refactor, add a way to disable mixed json and freestring

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* Fix linting issues

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

---------

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
2024-05-22 00:14:16 +02:00
Ettore Di Giacinto
a31d00d904
feat(aio): switch to llama3-based for LLM (#2225)
Signed-off-by: mudler <mudler@localai.io>
2024-05-03 00:41:45 +02:00
Ettore Di Giacinto
48d0aa2f6d
models(gallery): add new models to the gallery (#2124)
* models: add reranker and parler-tts-mini

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* fix: chatml im_end should not have a newline

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* models(noromaid): add

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* models(llama3): add 70b, add dolphin2.9

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* models(llama3): add unholy-8b

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

* models(llama3): add therapyllama3, aura

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

---------

Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
2024-04-25 01:28:02 +02:00
Ettore Di Giacinto
b664edde29
feat(rerankers): Add new backend, support jina rerankers API (#2121)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
2024-04-25 00:19:02 +02:00
Ettore Di Giacinto
b2772509b4
models(llama3): add llama3 to embedded models (#2074)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
2024-04-19 18:23:44 +02:00
Ettore Di Giacinto
f36d86ba6d
fix(hermes-2-pro-mistral): correct dashes in template to suppress newlines (#1966)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
2024-04-07 18:23:47 +02:00
Ettore Di Giacinto
84e0dc3246
fix(hermes-2-pro-mistral): correct stopwords (#1947)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
2024-04-02 15:38:00 +02:00
cryptk
86bc5f1350
fix: use exec in entrypoint scripts to fix signal handling (#1943) 2024-04-02 09:15:44 +02:00
Ettore Di Giacinto
ebb1fcedea
fix(hermes-2-pro-mistral): add stopword for toolcall (#1939)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
2024-04-01 11:48:35 +02:00
Ettore Di Giacinto
35290e146b
fix(grammar): respect JSONmode and grammar from user input (#1935)
* 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
2024-03-31 13:04:09 +02:00
Ettore Di Giacinto
957f428fd5
fix(tools): correctly render tools response in templates (#1932)
* fix(tools): allow to correctly display both Functions and Tools

* models(hermes-2-pro): correctly display function results
2024-03-30 19:02:07 +01:00
Ettore Di Giacinto
eab4a91a9b
fix(aio): correctly detect intel systems (#1931)
Also rename SIZE to PROFILE
2024-03-30 12:04:32 +01:00
Ettore Di Giacinto
e58410fa99
feat(aio): add intel profile (#1901)
* feat(aio): add intel profile

* docs: clarify AIO images features
2024-03-26 18:45:25 +01:00
Ettore Di Giacinto
c9adc5680c
fix(aio): make image-gen for GPU functional, update docs (#1895)
* readme: update quickstart

* aio(gpu): fix dreamshaper

* tests(aio): allow to run tests also against an endpoint

* docs: split content

* tests: less verbosity

---------

Co-authored-by: Dave <dave@gray101.com>
2024-03-25 21:04:32 +00:00
Enrico Ros
08c7b17298
Fix NVIDIA VRAM detection on WSL2 environments (#1894)
* NVIDIA VRAM detection on WSL2 environments

More robust single NVIDIA GPU memory detection, following the
improved NVIDIA WSL2 detection patch yesterday #1891.

Tested and working on WSL2, Linux.

Signed-off-by: Enrico Ros <enrico.ros@gmail.com>

* Update aio/entrypoint.sh

Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com>

---------

Signed-off-by: Enrico Ros <enrico.ros@gmail.com>
Signed-off-by: Ettore Di Giacinto <mudler@users.noreply.github.com>
Co-authored-by: Ettore Di Giacinto <mudler@users.noreply.github.com>
2024-03-25 18:36:18 +01:00
Enrico Ros
5e12382524
NVIDIA GPU detection support for WSL2 environments (#1891)
This change makes the assumption that "Microsoft Corporation Device 008e"
is an NVIDIA CUDA device. If this is not the case, please update the
hardware detection script here.

Signed-off-by: Enrico Ros <enrico.ros@gmail.com>
Co-authored-by: Dave <dave@gray101.com>
2024-03-25 08:32:40 +01:00
Ettore Di Giacinto
4b1ee0c170
feat(aio): add tests, update model definitions (#1880) 2024-03-22 21:13:11 +01:00
Ettore Di Giacinto
abc9360dc6
feat(aio): entrypoint, update workflows (#1872) 2024-03-21 22:09:04 +01:00
Ettore Di Giacinto
e533dcf506
feat(functions/aio): all-in-one images, function template enhancements (#1862)
* feat(startup): allow to specify models from local files

* feat(aio): add Dockerfile, make targets, aio profiles

* feat(template): add Function and LastMessage

* add hermes2-pro-mistral

* update hermes2 definition

* feat(template): add sprig

* feat(template): expose FunctionCall

* feat(aio): switch llm for text
2024-03-21 01:12:20 +01:00