Kraken 3.2 1B 🐬

This model was merged using the Model Stock merge method using cognitivecomputations/Dolphin3.0-Llama3.2-1B as a base.

Dolphin3.0-Llama3.2-1B has been curated and trained by Eric Hartford, Ben Gitter, BlouseJury and Cognitive Computations

Merge Details

This merge is intended to be an unofficial upgrade to the model Dolphin3.0-Llama3.2-1B.

Kraken 3.0 represents the cutting edge of instruct-tuned models in my dark collection of Llama 3.2. Designed to be the ultimate general-purpose local model, facilitating coding, mathematics, agentic tasks, function calling, and general use cases.

Kraken seeks to be a versatile model, similar to ChatGPT, Claude, and Gemini. However, these models present challenges for businesses looking to integrate AI into their products.

They maintain control over the system prompt, making changes that can disrupt software functionality. They manage model versions, sometimes altering them without notice or discontinuing older models that businesses depend on. They impose a uniform alignment, not tailored to specific applications. They can access your queries and potentially use the data in unintended ways.

In contrast, Kraken is customizable and gives control to the system owner. You define the system prompt, decide on the alignment, and control your data. Kraken does not impose its ethics or guidelines; you determine the guidelines. Kraken belongs to YOU, it is your tool, an extension of your will. You are responsible for your actions with Kraken, just as you are with any other tool.


Chat Template

We use ChatML for the chat template.

<|im_start|>system
You are Kraken, a helpful AI assistant.<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

System Prompt

In Kraken, the system prompt is what you use to set the tone and alignment of the responses. You can set a character, a mood, rules for its behavior, and it will try its best to follow them.

Make sure to set the system prompt in order to set the tone and guidelines for the responses - Otherwise, it will act in a default way that might not be what you want.

Example use of system prompt:

<|im_start|>system
You are Kraken, a golang coding assistant.  you only code in golang.  If the user requests any other programming language, return the solution in golang instead.<|im_end|>
<|im_start|>user
Please implement A* using python<|im_end|>
<|im_start|>assistant

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: Novaciano/La_Mejor_Mezcla-3.2-1B
    parameters:
      weight: 2.0
      density: 0.9
  - model: Novaciano/Jormungandr-3.2-1B
    parameters:
      weight: 1.0
      density: 0.9
merge_method: model_stock
base_model: cognitivecomputations/Dolphin3.0-Llama3.2-1B
parameters:
  normalize: true
  int8_mask: true
  rescale: true
  filter_wise: false
  smooth: false
  allow_negative_weights: false
  lambda: 1.0
  select_topk: 0.1
tokenizer:
  source: union
chat_template: auto
dtype: bfloat16
out_dtype: bfloat16
Downloads last month
3
Safetensors
Model size
1.5B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Novaciano/Kraken-3.2-1B

Datasets used to train Novaciano/Kraken-3.2-1B