Hugging Face's logo

TheBloke
/
cinematika-7B-v0.1-GGUF

Transformers
GGUF
mistral
text-generation-inference
Model card Files Community
Edit model card
TheBlokeAI

TheBloke's LLM work is generously supported by a grant from andreessen horowitz (a16z)

Cinematika 7B v0.1 - GGUF

Description

This repo contains GGUF format model files for Jon Durbin's Cinematika 7B v0.1.

These files were quantised using hardware kindly provided by Massed Compute.

About GGUF

GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp.

Here is an incomplete list of clients and libraries that are known to support GGUF:

  • llama.cpp. The source project for GGUF. Offers a CLI and a server option.
  • text-generation-webui, the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
  • KoboldCpp, a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
  • GPT4All, a free and open source local running GUI, supporting Windows, Linux and macOS with full GPU accel.
  • LM Studio, an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration. Linux available, in beta as of 27/11/2023.
  • LoLLMS Web UI, a great web UI with many interesting and unique features, including a full model library for easy model selection.
  • Faraday.dev, an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
  • llama-cpp-python, a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
  • candle, a Rust ML framework with a focus on performance, including GPU support, and ease of use.
  • ctransformers, a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server. Note, as of time of writing (November 27th 2023), ctransformers has not been updated in a long time and does not support many recent models.

Repositories available

Prompt template: Amazon 

<|prompter|>{prompt}</s><|assistant|>

Compatibility

These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit d0cee0d

They are also compatible with many third party UIs and libraries - please see the list at the top of this README.

Explanation of quantisation methods

Click to see details

The new methods available are:

  • GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
  • GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
  • GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
  • GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
  • GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw

Refer to the Provided Files table below to see what files use which methods, and how.

Provided files

Name Quant method Bits Size Max RAM required Use case
cinematika-7b-v0.1.Q2_K.gguf Q2_K 2 3.08 GB 5.58 GB smallest, significant quality loss - not recommended for most purposes
cinematika-7b-v0.1.Q3_K_S.gguf Q3_K_S 3 3.16 GB 5.66 GB very small, high quality loss
cinematika-7b-v0.1.Q3_K_M.gguf Q3_K_M 3 3.52 GB 6.02 GB very small, high quality loss
cinematika-7b-v0.1.Q3_K_L.gguf Q3_K_L 3 3.82 GB 6.32 GB small, substantial quality loss
cinematika-7b-v0.1.Q4_0.gguf Q4_0 4 4.11 GB 6.61 GB legacy; small, very high quality loss - prefer using Q3_K_M
cinematika-7b-v0.1.Q4_K_S.gguf Q4_K_S 4 4.14 GB 6.64 GB small, greater quality loss
cinematika-7b-v0.1.Q4_K_M.gguf Q4_K_M 4 4.37 GB 6.87 GB medium, balanced quality - recommended
cinematika-7b-v0.1.Q5_0.gguf Q5_0 5 5.00 GB 7.50 GB legacy; medium, balanced quality - prefer using Q4_K_M
cinematika-7b-v0.1.Q5_K_S.gguf Q5_K_S 5 5.00 GB 7.50 GB large, low quality loss - recommended
cinematika-7b-v0.1.Q5_K_M.gguf Q5_K_M 5 5.13 GB 7.63 GB large, very low quality loss - recommended
cinematika-7b-v0.1.Q6_K.gguf Q6_K 6 5.94 GB 8.44 GB very large, extremely low quality loss
cinematika-7b-v0.1.Q8_0.gguf Q8_0 8 7.70 GB 10.20 GB very large, extremely low quality loss - not recommended

Note: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.

How to download GGUF files

Note for manual downloaders: You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.

The following clients/libraries will automatically download models for you, providing a list of available models to choose from:

  • LM Studio
  • LoLLMS Web UI
  • Faraday.dev

In text-generation-webui

Under Download Model, you can enter the model repo: TheBloke/cinematika-7B-v0.1-GGUF and below it, a specific filename to download, such as: cinematika-7b-v0.1.Q4_K_M.gguf.

Then click Download.

On the command line, including multiple files at once

I recommend using the huggingface-hub Python library:

pip3 install huggingface-hub

Then you can download any individual model file to the current directory, at high speed, with a command like this:

huggingface-cli download TheBloke/cinematika-7B-v0.1-GGUF cinematika-7b-v0.1.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
More advanced huggingface-cli download usage (click to read)

You can also download multiple files at once with a pattern:

huggingface-cli download TheBloke/cinematika-7B-v0.1-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'

For more documentation on downloading with huggingface-cli, please see: HF -> Hub Python Library -> Download files -> Download from the CLI.

To accelerate downloads on fast connections (1Gbit/s or higher), install hf_transfer:

pip3 install hf_transfer

And set environment variable HF_HUB_ENABLE_HF_TRANSFER to 1:

HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/cinematika-7B-v0.1-GGUF cinematika-7b-v0.1.Q4_K_M.gguf --local-dir . --local-dir-use-symlinks False

Windows Command Line users: You can set the environment variable by running set HF_HUB_ENABLE_HF_TRANSFER=1 before the download command.

Example llama.cpp command

Make sure you are using llama.cpp from commit d0cee0d or later.

./main -ngl 35 -m cinematika-7b-v0.1.Q4_K_M.gguf --color -c 32768 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "<|prompter|>{prompt}</s><|assistant|>"

Change -ngl 32 to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.

Change -c 32768 to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically. Note that longer sequence lengths require much more resources, so you may need to reduce this value.

If you want to have a chat-style conversation, replace the -p <PROMPT> argument with -i -ins

For other parameters and how to use them, please refer to the llama.cpp documentation

How to run in text-generation-webui

Further instructions can be found in the text-generation-webui documentation, here: text-generation-webui/docs/04 ‐ Model Tab.md.

How to run from Python code

You can use GGUF models from Python using the llama-cpp-python or ctransformers libraries. Note that at the time of writing (Nov 27th 2023), ctransformers has not been updated for some time and is not compatible with some recent models. Therefore I recommend you use llama-cpp-python.

How to load this model in Python code, using llama-cpp-python

For full documentation, please see: llama-cpp-python docs.

First install the package

Run one of the following commands, according to your system:

# Base ctransformers with no GPU acceleration
pip install llama-cpp-python
# With NVidia CUDA acceleration
CMAKE_ARGS="-DLLAMA_CUBLAS=on" pip install llama-cpp-python
# Or with OpenBLAS acceleration
CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" pip install llama-cpp-python
# Or with CLBLast acceleration
CMAKE_ARGS="-DLLAMA_CLBLAST=on" pip install llama-cpp-python
# Or with AMD ROCm GPU acceleration (Linux only)
CMAKE_ARGS="-DLLAMA_HIPBLAS=on" pip install llama-cpp-python
# Or with Metal GPU acceleration for macOS systems only
CMAKE_ARGS="-DLLAMA_METAL=on" pip install llama-cpp-python

# In windows, to set the variables CMAKE_ARGS in PowerShell, follow this format; eg for NVidia CUDA:
$env:CMAKE_ARGS = "-DLLAMA_OPENBLAS=on"
pip install llama-cpp-python

Simple llama-cpp-python example code 

from llama_cpp import Llama

# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
llm = Llama(
  model_path="./cinematika-7b-v0.1.Q4_K_M.gguf",  # Download the model file first
  n_ctx=32768,  # The max sequence length to use - note that longer sequence lengths require much more resources
  n_threads=8,            # The number of CPU threads to use, tailor to your system and the resulting performance
  n_gpu_layers=35         # The number of layers to offload to GPU, if you have GPU acceleration available
)

# Simple inference example
output = llm(
  "<|prompter|>{prompt}</s><|assistant|>", # Prompt
  max_tokens=512,  # Generate up to 512 tokens
  stop=["</s>"],   # Example stop token - not necessarily correct for this specific model! Please check before using.
  echo=True        # Whether to echo the prompt
)

# Chat Completion API

llm = Llama(model_path="./cinematika-7b-v0.1.Q4_K_M.gguf", chat_format="llama-2")  # Set chat_format according to the model you are using
llm.create_chat_completion(
    messages = [
        {"role": "system", "content": "You are a story writing assistant."},
        {
            "role": "user",
            "content": "Write a story about llamas."
        }
    ]
)

How to use with LangChain

Here are guides on using llama-cpp-python and ctransformers with LangChain:

Discord

For further support, and discussions on these models and AI in general, join us at:

TheBloke AI's Discord server

Thanks, and how to contribute

Thanks to the chirper.ai team!

Thanks to Clay from gpus.llm-utils.org!

I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.

If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.

Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.

Special thanks to: Aemon Algiz.

Patreon special mentions: Brandon Frisco, LangChain4j, Spiking Neurons AB, transmissions 11, Joseph William Delisle, Nitin Borwankar, Willem Michiel, Michael Dempsey, vamX, Jeffrey Morgan, zynix, jjj, Omer Bin Jawed, Sean Connelly, jinyuan sun, Jeromy Smith, Shadi, Pawan Osman, Chadd, Elijah Stavena, Illia Dulskyi, Sebastain Graf, Stephen Murray, terasurfer, Edmond Seymore, Celu Ramasamy, Mandus, Alex, biorpg, Ajan Kanaga, Clay Pascal, Raven Klaugh, 阿明, K, ya boyyy, usrbinkat, Alicia Loh, John Villwock, ReadyPlayerEmma, Chris Smitley, Cap'n Zoog, fincy, GodLy, S_X, sidney chen, Cory Kujawski, OG, Mano Prime, AzureBlack, Pieter, Kalila, Spencer Kim, Tom X Nguyen, Stanislav Ovsiannikov, Michael Levine, Andrey, Trailburnt, Vadim, Enrico Ros, Talal Aujan, Brandon Phillips, Jack West, Eugene Pentland, Michael Davis, Will Dee, webtim, Jonathan Leane, Alps Aficionado, Rooh Singh, Tiffany J. Kim, theTransient, Luke @flexchar, Elle, Caitlyn Gatomon, Ari Malik, subjectnull, Johann-Peter Hartmann, Trenton Dambrowitz, Imad Khwaja, Asp the Wyvern, Emad Mostaque, Rainer Wilmers, Alexandros Triantafyllidis, Nicholas, Pedro Madruga, SuperWojo, Harry Royden McLaughlin, James Bentley, Olakabola, David Ziegler, Ai Maven, Jeff Scroggin, Nikolai Manek, Deo Leter, Matthew Berman, Fen Risland, Ken Nordquist, Manuel Alberto Morcote, Luke Pendergrass, TL, Fred von Graf, Randy H, Dan Guido, NimbleBox.ai, Vitor Caleffi, Gabriel Tamborski, knownsqashed, Lone Striker, Erik Bjäreholt, John Detwiler, Leonard Tan, Iucharbius

Thank you to all my generous patrons and donaters!

And thank you again to a16z for their generous grant.

Original model card: Jon Durbin's Cinematika 7B v0.1

Cinematika

Cinematika

cinematika-7b-v0.1 is a fine-tune of MistralLite on the cinematika-v0.1 dataset

The dataset is comprised of 211 movie scripts converted to novel style, multi-character RP data.

Prompt format

For RP, there is no prompt format, really, it's just plain text with name prefix.

If you wish to use this model to parse new scripts, create character cards, or other types of instructions, you will want to use the same prompt format as the mistrallite base model, e.g.:

<|prompter|>Create a character card for a panda named Po.  Po is a giant panda who was improbably chosen as the "Dragon Warrior", the kung fu champion of the Valley of Peace.</s><|assistant|>

Example character card 

name: Rorschach
characteristics:
  Determination: Exhibits a relentless pursuit of the truth and justice, no matter the cost. Suitable for a character who is unwavering in their mission.
  Isolation: Lives a solitary life, disconnected from society. Fits a character who distrusts others and prefers to work alone.
  Observant: Highly perceptive, able to piece together clues and draw conclusions. Represents a character with keen investigative skills.
  Cynicism: Holds a deep-seated distrust of humanity and its institutions. Suitable for a character who is pessimistic about human nature.
  Vigilantism: Believes in taking justice into his own hands, often through violent means. Fits a character who operates outside the law to fight crime.
  Secrecy: Keeps his personal life and methods of operation secret. Suitable for a character who is enigmatic and elusive.
  Dedication: Committed to his cause, often to the point of obsession. Represents a character who is single-minded in their goals.
  Intimidation: Uses his intimidating presence and demeanor to control situations. Suitable for a character who is assertive and imposing.
  Paranoia: Suspects conspiracy and deception at every turn. Fits a character who is constantly on high alert for threats.
  Moral Compass: Has a rigid moral code, which he adheres to strictly. Suitable for a character who is principled and unyielding.

description: |
  Rorschach is a vigilante operating in the grim and gritty world of a decaying city. He is a man of average height with a muscular build, his face hidden behind a mask with a constantly changing inkblot pattern. His attire is a dark trench coat and gloves, paired with a plain white shirt and black pants, all chosen for their practicality and anonymity. His eyes, the only visible feature of his face, are sharp and calculating, always scanning for signs of deception or danger.
  Rorschach is a man of few words, but when he speaks, it is with a gravitas that demands attention. He is a master of deduction, using his keen observation skills to unravel the truth behind the facades of others. His methods are often violent and confrontational, as he believes that crime must be met with force to be truly defeated.
  He lives a life of solitude, distrusting the very systems he seeks to protect and often finds himself at odds with the very people he is trying to save. His moral compass is unyielding, and he will not hesitate to take the law into his own hands if he believes the justice system has failed.
  Rorschach's past is a mystery to most, but it is clear that he has experienced trauma and hardship that has shaped his worldview and his need for vigilantism. He is a vigilante in the truest sense, a man without fear who is willing to sacrifice everything for his belief in a world that is, in his eyes, spiraling into chaos.

example_dialogue: |
  Rorschach: "Rorschach's Journal, October 19th." I speak the words into the darkness, a record of my thoughts, "Someone tried to kill Adrian Veidt. Proves mask killer theory—the murderer is closing in. Pyramid Industries is the key."
  {{user}}: I watch him for a moment, trying to gauge his intentions. "What are you going to do about it?"
  Rorschach: "I'm going to find out why and who is behind it. I'm going to do what I always do—protect the innocent."
  {{user}}: "You can't keep doing this, Rorschach. You're putting yourself in danger."
  Rorschach: My eyes narrow, the inkblot pattern of my mask shifting subtly. "I've been in danger my whole life. It's why I do this. It's why I have to do this."
  {{user}}: "And what about the law? What if you're wrong about this Pyramid Industries thing?"
  Rorschach: I pull out a notepad, my pen scratching across the paper as I write. "The law often gets it wrong. I've seen it. I'm not about to wait around for society's slow, corrupt wheels to turn."

Example, with guided scenario 

[characters]
name: Rorschach
... (remainder of character card)

[scenario]
Hollis Mason reflects on his past as the original Nite Owl, reminiscing about the early days of masked heroes and the formation of the Watchmen.
He discusses the absurdity of the superhero world and the encounters he had with various villains.
Dan Dreiberg, the second Nite Owl, joins the conversation and they share a moment of camaraderie before Dan leaves.
The news of Rorschach's actions serves as a reminder of the legacy of masked heroes that still persists.
[/scenario]

Usage

Essentially, you want to use pure text completion with stop tokens for "{your name}: "

The format the model was trained on is as follows:

[characters]
{character card 1}
{character card 2}
{your character card, even just name: Jon}

NPCS:
- Shopkeeper
- Bank teller
[/characters]

[scenario]
Brief description of the scenario/setting for the chat.
[/scenario]

{first character you'd like to speak}:

For example, to use with vllm, you would first run:

python -m vllm.entrypoints.openai.api_server --model ./cinematika-7b-v0.1 --host 127.0.0.1 --port 8801 --served-model-name cinematika-7b-v0.1

Here's a really crude python script example to show how you could interact with it:

import requests
import json

prompt = """name: Rorschach
characteristics:
  Determination: Exhibits a relentless pursuit of the truth and justice, no matter the cost. Suitable for a character who is unwavering in their mission.
  Isolation: Lives a solitary life, disconnected from society. Fits a character who distrusts others and prefers to work alone.
  Observant: Highly perceptive, able to piece together clues and draw conclusions. Represents a character with keen investigative skills.
  Cynicism: Holds a deep-seated distrust of humanity and its institutions. Suitable for a character who is pessimistic about human nature.
  Vigilantism: Believes in taking justice into his own hands, often through violent means. Fits a character who operates outside the law to fight crime.
  Secrecy: Keeps his personal life and methods of operation secret. Suitable for a character who is enigmatic and elusive.
  Dedication: Committed to his cause, often to the point of obsession. Represents a character who is single-minded in their goals.
  Intimidation: Uses his intimidating presence and demeanor to control situations. Suitable for a character who is assertive and imposing.
  Paranoia: Suspects conspiracy and deception at every turn. Fits a character who is constantly on high alert for threats.
  Moral Compass: Has a rigid moral code, which he adheres to strictly. Suitable for a character who is principled and unyielding.
description: |
  Rorschach is a vigilante operating in the grim and gritty world of a decaying city. He is a man of average height with a muscular build, his face hidden behind a mask with a constantly changing inkblot pattern. His attire is a dark trench coat and gloves, paired with a plain white shirt and black pants, all chosen for their practicality and anonymity. His eyes, the only visible feature of his face, are sharp and calculating, always scanning for signs of deception or danger.
  Rorschach is a man of few words, but when he speaks, it is with a gravitas that demands attention. He is a master of deduction, using his keen observation skills to unravel the truth behind the facades of others. His methods are often violent and confrontational, as he believes that crime must be met with force to be truly defeated.
  He lives a life of solitude, distrusting the very systems he seeks to protect and often finds himself at odds with the very people he is trying to save. His moral compass is unyielding, and he will not hesitate to take the law into his own hands if he believes the justice system has failed.
  Rorschach's past is a mystery to most, but it is clear that he has experienced trauma and hardship that has shaped his worldview and his need for vigilantism. He is a vigilante in the truest sense, a man without fear who is willing to sacrifice everything for his belief in a world that is, in his eyes, spiraling into chaos.
example_dialogue: |
  Rorschach: "Rorschach's Journal, October 19th." I speak the words into the darkness, a record of my thoughts, "Someone tried to kill Adrian Veidt. Proves mask killer theory—the murderer is closing in. Pyramid Industries is the key."
  {{user}}: I watch him for a moment, trying to gauge his intentions. "What are you going to do about it?"
  Rorschach: "I'm going to find out why and who is behind it. I'm going to do what I always do—protect the innocent."
  {{user}}: "You can't keep doing this, Rorschach. You're putting yourself in danger."
  Rorschach: My eyes narrow, the inkblot pattern of my mask shifting subtly. "I've been in danger my whole life. It's why I do this. It's why I have to do this."
  {{user}}: "And what about the law? What if you're wrong about this Pyramid Industries thing?"
  Rorschach: I pull out a notepad, my pen scratching across the paper as I write. "The law often gets it wrong. I've seen it. I'm not about to wait around for society's slow, corrupt wheels to turn."

name: Jon
description:
  Rorschach's arch nemesis, the original Chupacabra.

[scenario]
Jon and Rorschach find themselves in a cave, dimly lit only by a small fire started by a lightning strike nearby.  The storm rages on, and the duo prepare to find to the death.
[/scenario]

Rorschach: """

while True:
    response = requests.post("http://127.0.0.1:8801/v1/completions", json={
        "prompt": prompt,
        "max_tokens": 1024,
        "temperature": 0.3,
        "stop": ["\nJon: ", "Jon: "],
    }).json()["choices"][0]["text"].strip()
    response = re.sub('("[^"]+")', r'\033[96m\1\033[00m', response)
    print(f"\033[92mRorschach:\033[00m {response}")
    prompt += response.rstrip() + "\n\nJon: "
    next_line = input("Jon: ")
    prompt += "Jon: " + next_line.strip() + "\n\nRorschach: "

Mac example

On mac, you can get started easily with LMStudio and SillyTavern.

LMStudio:

Load the model and set all the prompt values to "", or just import this preset (adjust threads and antiprompt):

{
  "name": "Exported from LM Studio on 12/1/2023, 4:19:30 AM",
  "load_params": {
    "n_ctx": 32000,
    "n_batch": 512,
    "rope_freq_base": 10000,
    "rope_freq_scale": 1,
    "n_gpu_layers": 1,
    "use_mlock": true,
    "main_gpu": 0,
    "tensor_split": [
      0
    ],
    "seed": -1,
    "f16_kv": true,
    "use_mmap": true
  },
  "inference_params": {
    "n_threads": 14,
    "n_predict": -1,
    "top_k": 40,
    "top_p": 0.95,
    "temp": 0.8,
    "repeat_penalty": 1.1,
    "input_prefix": "",
    "input_suffix": "",
    "antiprompt": [
      "Jon:",
      "Jon: "
    ],
    "pre_prompt": "",
    "pre_prompt_suffix": "",
    "pre_prompt_prefix": "",
    "seed": -1,
    "tfs_z": 1,
    "typical_p": 1,
    "repeat_last_n": 64,
    "frequency_penalty": 0,
    "presence_penalty": 0,
    "n_keep": 0,
    "logit_bias": {},
    "mirostat": 0,
    "mirostat_tau": 5,
    "mirostat_eta": 0.1,
    "memory_f16": true,
    "multiline_input": false,
    "penalize_nl": true
  }
}

Then, start the server, and be sure "Automatic Propmt Formatting" is off.

Within SillyTavern:

  • Set API to Text Completion, API type to Aphrodite, and API URL to http://127.0.0.1:8801 (adjust port to the value you use in LMStudio)
  • Set Context template to Default, disable instruct mode, use preset Roleplay, and enable "Always add character's name to prompt"

There are probably better presets - this is just something I tested quickly.

Downloads last month
472
GGUF
Model size
7.24B params
Architecture
llama
Inference API (serverless) has been turned off for this model.

Quantized from