LLM EXPLORER 59,358 MODELS INDEXED

TeenyTinyLlama 460M Chat AWQ by nicholasKluge

By nicholasKluge · 4 downloads
HF Repo ↗ Copy URL

TeenyTinyLlama 460M Chat AWQ is an open-source language model by nicholasKluge. Features: 460m LLM, VRAM: 0.3GB, Context: 2K, License: apache-2.0, Quantized, Instruction-Based, LLM Explorer Score: 0.11.

  Arxiv:2401.16640   4-bit   Alignment   Assistant   Awq Base model:nicholaskluge/teeny... Base model:quantized:nicholask...   Co2 eq emissions   Conversation   Conversational Dataset:nicholaskluge/instruct...   Instruct   Llama   Pt   Quantized   Region:us   Safetensors
Model Card on HF 🤗: https://huggingface.co/nicholasKluge/TeenyTinyLlama-460m-Chat-awq 

TeenyTinyLlama 460M Chat AWQ Parameters and Internals

Model Type 
text generation, conversation, assistant
Use Cases 
Areas:
Research on low-resource language modeling
Applications:
Scientific experiments, Fine-tuning for deployment
Primary Use Cases:
Understand challenges related to developing language models in low-resource languages
Limitations:
Not intended for deployment, Not suitable for human-facing interactions, Not suitable for translation or generating text in other languages
Considerations:
Users are warned of risks, biases, and the need for human moderation.
Additional Notes 
Use CAUTION in deploying as it inherits general language model biases and behaviors.
Supported Languages 
Brazilian Portuguese (high)
Training Details 
Data Sources:
nicholasKluge/instruct-aira-dataset-v2
Data Volume:
unknown
Methodology:
fine-tuning
Hardware Used:
NVIDIA A100-SXM4-40GB
Model Architecture:
quantized version using AutoAWQ
Responsible Ai Considerations 
Fairness:
The model can produce toxic content, i.e., harmful, offensive, or detrimental to individuals, groups, or communities.
Transparency:
Users are encouraged to perform their risk analysis on these models.
Accountability:
Developers of TeenyTinyLlama
Mitigation Strategies:
Hasn't been outlined specifically.
Input Output 
Input Format:
Uses special tokens to demarcate user queries and model responses.
Accepted Modalities:
text
Output Format:
Generates text in response to queries.
Performance Tips:
Ensure human moderation in interactive applications.
LLM NameTeenyTinyLlama 460M Chat AWQ
Repository 🤗https://huggingface.co/nicholasKluge/TeenyTinyLlama-460m-Chat-awq 
Base Model(s)  TeenyTinyLlama 460M   nicholasKluge/TeenyTinyLlama-460m
Model Size460m
Required VRAM0.3 GB
Updated2026-07-14
MaintainernicholasKluge
Model Typellama
Instruction-BasedYes
Model Files  0.3 GB
Supported Languagespt
AWQ QuantizationYes
Quantization Typeawq
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length2048
Model Max Length2048
Transformers Version4.35.2
Tokenizer ClassLlamaTokenizer
Padding Token<pad>
Vocabulary Size32002
Torch Data Typefloat16

Best Alternatives to TeenyTinyLlama 460M Chat AWQ

Best Alternatives
Context / RAM
Downloads
Likes
TeenyTinyLlama 460M AWQ2K / 0.3 GB31
TeenyTinyLlama 460M Chat2K / 0 GB903
TeenyTinyLlama 460M2K / 1.9 GB14512
...60M Experimental Ptbr Instruct2K / 0.9 GB84
Note: green Score (e.g. "73.2") means that the model is better than nicholasKluge/TeenyTinyLlama-460m-Chat-awq.