LLM EXPLORER 59,420 MODELS INDEXED

Stablelm Tuned Alpha 3B by stabilityai

By stabilityai · 746 downloads

Stablelm Tuned Alpha 3B is an open-source language model by stabilityai. Features: 3b LLM, VRAM: 14.9GB, Context: 4K, License: cc-by-nc-sa-4.0, LLM Explorer Score: 0.12, Arc: 27.8, HellaSwag: 44.1, MMLU: 23.1, GSM8K: 0.5.

  Dataset:dahoas/full-hh-rlhf   Dataset:dmayhem93/chatcombined Dataset:huggingfaceh4/databric... Dataset:jeffwan/sharegpt vicun... Dataset:nomic-ai/gpt4all promp...   Dataset:tatsu-lab/alpaca   Deploy:azure   En   Endpoints compatible   Gpt neox   Pytorch   Region:us   Sharded

Stablelm Tuned Alpha 3B Parameters and Internals

Model Type 
causal-lm
Use Cases 
Areas:
open-source community, chat-like applications
Limitations:
The model may generate biased or toxic text despite efforts in safe fine-tuning., Not intended as a replacement for human judgment
Considerations:
Be mindful of potential bias or toxic outputs.
Additional Notes 
Models include a helpful hand from Dakota Mahan ([@dmayhem93](https://huggingface.co/dmayhem93)) in their development.
Supported Languages 
English (Proficient)
Training Details 
Data Sources:
tatsu-lab/alpaca, nomic-ai/gpt4all_prompt_generations, Dahoas/full-hh-rlhf, jeffwan/sharegpt_vicuna, HuggingFaceH4/databricks_dolly_15k
Methodology:
Supervised fine-tuning on natural language datasets focused on chat and instruction-following tasks.
Context Length:
4096
Model Architecture:
NeoX transformer architecture
Responsible Ai Considerations 
Fairness:
Models are developed to adhere to safer distributions of text but cannot mitigate all biases and toxicity.
Transparency:
It should not be treated as a substitute for human judgment or considered a source of truth.
Accountability:
Users are responsible for the outputs generated and should use models responsibly.
Mitigation Strategies:
Fine-tuning on datasets aimed at improving safety, but may not remove all biases/toxicity.
Input Output 
Input Format:
Prompts formatted to <|SYSTEM|>...<|USER|>...<|ASSISTANT|>...
Accepted Modalities:
text
Output Format:
Text output
LLM NameStablelm Tuned Alpha 3B
Repository πŸ€—https://huggingface.co/stabilityai/stablelm-tuned-alpha-3b 
Model Size3b
Required VRAM14.9 GB
Updated2026-07-09
Maintainerstabilityai
Model Typegpt_neox
Model Files  10.2 GB: 1-of-2   4.7 GB: 2-of-2
Supported Languagesen
Model ArchitectureGPTNeoXForCausalLM
Licensecc-by-nc-sa-4.0
Context Length4096
Model Max Length4096
Transformers Version4.28.1
Tokenizer ClassGPTNeoXTokenizer
Vocabulary Size50688
Torch Data Typefloat32

Quantized Models of the Stablelm Tuned Alpha 3B

Model
Likes
Downloads
VRAM
Stablelm Tuned Alpha 3B 8bit3144 GB
Stablelm Tuned Alpha 3B 16bit6137 GB

Best Alternatives to Stablelm Tuned Alpha 3B

Best Alternatives
Context / RAM
Downloads
Likes
Stablecode Completion Alpha 3B16K / 14.1 GB180119
RedPajama 3B 1638416K / 19.7 GB104
Redpajama 3B Chat5K / 6.4 GB132
Stablelm Base Alpha 3B4K / 14.9 GB313982
...blecode Completion Alpha 3B 4K4K / 6.1 GB434278
Stablecode Instruct Alpha 3B4K / 6.1 GB2304
StableCode 3B4K / 6.1 GB81
...tion Alpha 3B 4K Openvino Int84K / 2.8 GB81
Redpajama 3B Evol Coder4K / 6.1 GB11
Literature 3B 40964K / 11.7 GB186
Note: green Score (e.g. "73.2") means that the model is better than stabilityai/stablelm-tuned-alpha-3b.