-
This seems to work after a long long time
```bash
# Command
curl http://127.0.0.1:8080/v1/chat/completions -H "Content-Type: application/json" -d '{ "messages": [{ "role": "user", "content": "tel…
-
Greetings!
I run your code 'examples/exploration_task_tree.py' and get a mp4 which shows how the arm pick and put a crayon box. And i try to figure out where is the query.'
In your 'docs/walkthrough…
-
微博内容精选
-
### Feature Description
- [ ] Retrieval: https://github.com/jerryjliu/llama_index/issues/6554
- [ ] RAG / Knowledge-Intensive QA: https://github.com/jerryjliu/llama_index/issues/6919
- [ ] MMLU?
…
-
Hi Folks, Thank you for the awesome library.
1 - What happens when we call Bumblebee.load_model(repository) ? It downloads the entire model to somewhere in my computer? Where?
2 - How to use a…
-
### Feature Description
Multistep queries are great. But they can pollute the context window and cause the query path to be prematurely commited to one strategy.
As a more general paradigm, mult…
-
Hi! I see you conduct the ablation experiments of Intensity-based augmentations and CutMix-based augmentations. But I wonder if there still exists obvious improvement between vanilla CutMix strategy a…
-
## 🐛 Bug
### Background context
I am successfully running `Llama-2-7b-chat-hf-q4f16_1` on a single AMD Instinct MI100 after building TVM from source as instructed in this comment https://github.c…
-
We are working on Haystack 2.0, with a major refactoring of pipelines and components.
## Rationale
We need to prioritize the list of components and separately the list of document stores to migrat…
-
Hello. I would like to request a feature for selecting or highlighting content.
Let us consider the following prompt:
```
You are a helpful assistant.
Given the following text:
Lorem ipsu…