Instructions to use TheBloke/Nous-Capybara-34B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use TheBloke/Nous-Capybara-34B-GGUF with Transformers:

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("TheBloke/Nous-Capybara-34B-GGUF", dtype="auto")

llama-cpp-python

How to use TheBloke/Nous-Capybara-34B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="TheBloke/Nous-Capybara-34B-GGUF",
	filename="nous-capybara-34b.Q2_K.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use TheBloke/Nous-Capybara-34B-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf TheBloke/Nous-Capybara-34B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf TheBloke/Nous-Capybara-34B-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf TheBloke/Nous-Capybara-34B-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf TheBloke/Nous-Capybara-34B-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf TheBloke/Nous-Capybara-34B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf TheBloke/Nous-Capybara-34B-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf TheBloke/Nous-Capybara-34B-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf TheBloke/Nous-Capybara-34B-GGUF:Q4_K_M

Use Docker

docker model run hf.co/TheBloke/Nous-Capybara-34B-GGUF:Q4_K_M

LM Studio
Jan
Ollama
How to use TheBloke/Nous-Capybara-34B-GGUF with Ollama:
```
ollama run hf.co/TheBloke/Nous-Capybara-34B-GGUF:Q4_K_M
```

Unsloth Studio new

How to use TheBloke/Nous-Capybara-34B-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for TheBloke/Nous-Capybara-34B-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for TheBloke/Nous-Capybara-34B-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for TheBloke/Nous-Capybara-34B-GGUF to start chatting

Docker Model Runner
How to use TheBloke/Nous-Capybara-34B-GGUF with Docker Model Runner:
```
docker model run hf.co/TheBloke/Nous-Capybara-34B-GGUF:Q4_K_M
```

Lemonade

How to use TheBloke/Nous-Capybara-34B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull TheBloke/Nous-Capybara-34B-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Nous-Capybara-34B-GGUF-Q4_K_M

List all available models

lemonade list

Incorrect EOS token

by Henk717 - opened Nov 14, 2023

Discussion

Henk717

Nov 14, 2023

•

edited Nov 14, 2023

Using Koboldcpp the model reports <|endoftext|> as the Endoftext token, however the finetune generates </s> as the end of text token.
This makes our usual techniques to produce nice outputs fail on this model, might be an upstream issue but the GGUF is definately not doing it right.

mljxy

Nov 14, 2023

upstream issue. The generated </s> is not from a single token.

Henk717

Nov 14, 2023

Reported it to them to, but if its not from a single token thats probably going to require a full retune.

wolfram

Nov 14, 2023

Yes, noticed that as well. SillyTavern is filtering it out, but I still am seeing it pop up before it gets deleted.

TheBloke

Owner Nov 14, 2023

Yeah I noticed that rogue </s> at the end of llama.cpp generation as well. It stops at the right point, but that extra token is there at the end.

I was going to suggest that we could use Kerfuffle's new script to set the GGUF EOS token to </s> but if you're saying it's not a single token ID, then I guess we can't?

KerfuffleV2

Nov 14, 2023

If you don't actually want the model to general HTML/code then you could possibly try setting logit biases that ban tokens that start with <. If it can't produce the weird </s> thing it might generate an EOS. If it just isn't generating EOS correctly, that won't help (but it will get rid of stray </s> in the output possibly. The base Yi model token id for </ is 1359 so with llama.cpp at least you can specify something like -l 1359-inf to ban the token.

Henk717

Nov 16, 2023

•

edited Nov 16, 2023

Workaround for Koboldcpp users is to use </s> as your stop sequence, problem is that you can't tell the model to keep generating that way so its short responses only.

phooney

Jan 2, 2024

Is there any update on this one? I'm using the GPTQ model and EXLLama2 on Kobold United and it has the same issue, but I think I can only use stop sequences when using it via the API? Any workaround for using as a stop sequence within the UI itself?

Henk717

Jan 3, 2024

Lite has it, and in the normal UI you might be having more luck with phrase biasing. At this point its safe to say the model won't be fixed and people should move on to their hermes or a different tune instead.

phooney

Jan 3, 2024

Ah, that's a shame, this seems like such a smart model at the size otherwise.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment