LLM EXPLORER
 LLM Explorer: A Curated Large Language Model Directory and Analytics
Dark Theme
  LLM  List    LLM Hosting    LLM Leaderboards    Blog    Newsfeed    Advertise

Granite 3.0 3B A800m Instruct by ibm-granite

 »  All LLMs  »  ibm-granite  »  Granite 3.0 3B A800m Instruct   URL Share it on

  1678   
  Arxiv:0000.00000   Autotrain compatible Base model:finetune:ibm-granit... Base model:ibm-granite/granite...   Conversational   Granite-3.0   Granitemoe   Instruct   Language   Model-index   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/ibm-granite/granite-3.0-3b-a800m-instruct 

Granite 3.0 3B A800m Instruct Benchmarks

MMLU Pro: 12.8
GPQA: 4.14
MUSR: 2.08
BBH: 13.16
IFEval: 42.98 vs 88 (so35)-51.2%
MATH Lvl 5: 7.02
LLME Score: 0.2424
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Granite 3.0 3B A800m Instruct (ibm-granite/granite-3.0-3b-a800m-instruct)
🌟 Advertise your project 🚀

Granite 3.0 3B A800m Instruct Parameters and Internals

Model Type 
text generation, instruction following
Use Cases 
Areas:
general instructions, AI assistants, business applications
Applications:
text generation, instruction following
Primary Use Cases:
Summarization, Text classification, Text extraction, Question-answering, Retrieval Augmented Generation (RAG), Code related tasks, Function-calling tasks, Multilingual dialog use cases
Limitations:
Might not perform equally across all languages as in English., Potential for inaccurate, biased, or unsafe responses without proper safety testing.
Considerations:
Proper safety testing and example tuning tailored for specific tasks.
Additional Notes 
The model infrastructure is environmentally friendly, leveraging 100% renewable energy.
Supported Languages 
English (supported), German (supported), Spanish (supported), French (supported), Japanese (supported), Portuguese (supported), Arabic (supported), Czech (supported), Italian (supported), Korean (supported), Dutch (supported), Chinese (supported)
Training Details 
Data Sources:
publicly available datasets with permissive license, internal synthetic data, human-curated data
Methodology:
supervised finetuning, model alignment using reinforcement learning, and model merging
Context Length:
4096
Hardware Used:
IBM's supercomputing cluster, Blue Vela with NVIDIA H100 GPUs
Model Architecture:
decoder-only sparse Mixture of Experts (MoE) transformer architecture
Responsible Ai Considerations 
Fairness:
multilingual data, but primary tuning on English instruction-response pairs.
Transparency:
Model developed by Granite Team, IBM. See accompanying technical documentation.
Mitigation Strategies:
Introducing few-shot learning for improved accuracy on multilingual tasks.
Input Output 
Input Format:
chat template with role, content fields
Accepted Modalities:
text
Output Format:
text
Performance Tips:
Adjust sequence length as required.
Release Notes 
Date:
October 21st, 2024
Notes:
Initial release with instruction tuning and multilingual capabilities.
LLM NameGranite 3.0 3B A800m Instruct
Repository 🤗https://huggingface.co/ibm-granite/granite-3.0-3b-a800m-instruct 
Base Model(s)  ibm-granite/granite-3.0-3b-a800m-base   ibm-granite/granite-3.0-3b-a800m-base
Model Size3b
Required VRAM6.8 GB
Updated2025-10-23
Maintaineribm-granite
Model Typegranitemoe
Instruction-BasedYes
Model Files  5.0 GB: 1-of-2   1.8 GB: 2-of-2
Model ArchitectureGraniteMoeForCausalLM
Licenseapache-2.0
Context Length4096
Model Max Length4096
Transformers Version4.46.0.dev0
Tokenizer ClassGPT2Tokenizer
Padding Token<|end_of_text|>
Vocabulary Size49155
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to Granite 3.0 3B A800m Instruct

Best Alternatives
Context / RAM
Downloads
Likes
Granite 3.1 3B A800m Instruct128K / 6.6 GB1346227
Note: green Score (e.g. "73.2") means that the model is better than ibm-granite/granite-3.0-3b-a800m-instruct.

Rank the Granite 3.0 3B A800m Instruct Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation  
Factuality and Completeness of Knowledge  
Censorship and Alignment  
Data Analysis and Insight Generation  
Text Generation  
Text Summarization and Feature Extraction  
Code Generation  
Multi-Language Support and Translation  

What open-source LLMs or SLMs are you in search of? 51545 in total.

Email us: info@extractum.io. Our Privacy Policy  | Terms and Conditions  | Suggest an improvement.
Our Social Media →  
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124
  Support LLM Explorer