Comments (4)
We can use mistral 7b to test this. It will also be useful to add a Langchain example that leverage this API.
from llm-on-ray.
@xuechendi is working on it.
from llm-on-ray.
@carsonwang, I've been working on this issue but a little lost on How to enable.
Observations:
-
HuggingFace support for 'tools', 'tool_choice': no native parameter support in generate function for "tools", "tool_choice". Seems huggingface has their own way to support Tools: https://huggingface.co/docs/transformers/main/en/custom_tools
-
RayLLM tools/tool_choice support. RayLLM supports these two keywords in client API, but after digging into inference codes, can't find how these two keywords passing to model.inference function either vLLMEngine or TRTEngine.
Even though AnyScale blog mentioned their fully support for function_calls, it is not come to RayLLM codes yet. Anyscale Endpoints: JSON Mode and Function calling Features -
llama_cpp toos/tool_choice support. I find function_call implementation in llama_cpp_python for two models: functionary and chatML. But only functionary is accepting these two keywords natually, chatML is still convert it as plain text as Langchain.initialize_agent did. https://github.com/abetlen/llama-cpp-python/blob/main/llama_cpp/llama_chat_format.py#L2032
-
vLLM / triton tensor-RT / huggingface issues research on tools/tool_chain support. Not found aligned conclusion.
Below is my branch for duplicating RayLLM supports for tools/tool_choice:
xuechendi@42d9613
from llm-on-ray.
pr: #134
from llm-on-ray.
Related Issues (20)
- Add serve command line options to list all supported model-ids (configured in *yaml)
- Issue about using ipex on cpu
- Migrate OpenAI API to 1.0 HOT 1
- Consolidate deepspeed workers for DeepSpeedPredictor and HPUPredictor
- Docker files for both CI and User
- Define simple_protocol.py and define pydantic SimpleRequest and SimpleModelResponse classes to encapsulate current json format
- Output some debug info in CI when Internal Server Error
- Calculate correct input length for every prompt in a single batch
- expected scalar type BFloat16 but found Float HOT 1
- Revise README.md in examples directory
- Build docker files for both CI and User
- Migrate CI to miniforge instead of miniconda
- A question about finetune dataset processing HOT 1
- Finetuning on Ray and CPU causes Runtime error HOT 12
- [Finetune] Fix fault-tolerant training
- [Finetune] WandB integration
- Inference Qwen1.5-110b on Gaudi2 HOT 1
- Inference Mixtral on Gaudi HOT 3
- Finetune on Ray cluster with trial task failed HOT 4
- AssertionError: BF16 weight prepack needs the cpu support avx512bw, avx512vl and avx512dq HOT 5
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from llm-on-ray.