LLM EXPLORER 59,420 MODELS INDEXED

Carrot Ko 2B Instruct by CarrotAI

By CarrotAI · 17 downloads

Carrot Ko 2B Instruct is an open-source language model by CarrotAI. Features: 2b LLM, VRAM: 4.2GB, Context: 32K, License: mit, Instruction-Based, LLM Explorer Score: 0.13.

  Conversational Dataset:carrotai/ko-instructio...   Endpoints compatible   Instruct   Ko   Pytorch   Qwen2   Region:us

Carrot Ko 2B Instruct Parameters and Internals

Use Cases 
Applications:
chatbots, question-answering systems
Limitations:
Performance can vary based on task complexity and input specifics.
Considerations:
Users should evaluate the model thoroughly for specific contexts.
Additional Notes 
This model is still experimental and its performance is being evaluated.
Supported Languages 
ko (Korean)
Training Details 
Data Sources:
CarrotAI/ko-instruction-dataset
Methodology:
Fine-tuned using axolotl.
LLM NameCarrot Ko 2B Instruct
Repository πŸ€—https://huggingface.co/CarrotAI/Carrot-Ko-2B-Instruct 
Model Size2b
Required VRAM4.2 GB
Updated2026-08-25
MaintainerCarrotAI
Model Typeqwen2
Instruction-BasedYes
Model Files  4.2 GB
Supported Languagesko
Model ArchitectureQwen2ForCausalLM
Licensemit
Context Length32768
Model Max Length32768
Transformers Version4.42.3
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size151936
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to Carrot Ko 2B Instruct

Best Alternatives
Context / RAM
Downloads
Likes
OpenRHO 2B Thinker128K / 3.5 GB72
Qwen2.5 2B Instruct32K / 5.9 GB121
Qwen2 Math 2B Instruct4K / 4.3 GB50
Note: green Score (e.g. "73.2") means that the model is better than CarrotAI/Carrot-Ko-2B-Instruct.