-
Changing ENV TRANSFORMERS_CACHE=/workspace/.cache/huggingface/hub/ to huggingface/transformers/ causes all models already in ~/.cache/huggingface/hub to be downloaded again into ~/.cache/huggingface/t…
-
When running behind corporate firewall and running using either cli
``` sgpt --code "test"```
i get a long python error about certificats
here is a screenshot of the ending
![image](https://github…
-
### Is your feature request related to a problem? Please describe.
大神好,遇到的问题还是较长的文本还有些表意问题,包括未来如果加入prompt来做similarity search,肯定需要foundation model很强。
### Solutions
现在看下来认真用较大比例中文语料搞的开源的,刚出来的ba…
-
Hey :wave:
I'm trying to package this tool for Arch Linux, but our standard way of packaging PEP 517 compliant projects produces an empty wheel. I believe this is because of a missing line in `pypr…
-
The `/run` command is useful in the chatbox, but it just run exactly what I type a shell script.
It would be super useful(highly expected) to just translate what is in front of /run into a shell scri…
-
Hello,
i am in syria and this country is banned so i want to change the endpoint so how?
thanks in advance.
-
I get this error whenever I use sgpt.
```
sgpt "mass of sun"
```
```
╭─────────────────────────────── Traceback (most recent call last) ────────────────────────────────╮
│ /home/tedwards/mamba…
-
FIRSTLY, AWESOME REPO, LOVE IT!!!
Since llama.cpp/llama-cpp-python can run a huge number of llama etc models on an API that basically mimics OpenAI's API, we can run sgpt for free with any model
…
-
An error occurs when running sgpt:
```
~> sgpt what is my ip …
-
I am on version 0.9.1
Everything is working fine, except for gpt-4.
Prompt: ```sgpt --model "gpt-4" "Does this work?"```
Output:
```
╭─────────────────────────────── Traceback (most recen…