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

Jamba V0.1 by ai21labs

 »  All LLMs  »  ai21labs  »  Jamba V0.1   URL Share it on

  2896   
  Arxiv:2403.19887   Autotrain compatible   Custom code   Endpoints compatible   Jamba   Mamba   Moe   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/ai21labs/Jamba-v0.1 

Jamba V0.1 Benchmarks

MMLU Pro: 16.57
GPQA: 2.46
MUSR: 3.71
BBH: 10.72
IFEval: 20.26 vs 88 (so35)-77%
MATH Lvl 5: 1.59
LLME Score: 0.26386
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Jamba V0.1 (ai21labs/Jamba-v0.1)
🌟 Advertise your project 🚀

Jamba V0.1 Parameters and Internals

Model Type 
Hybrid SSM-Transformer, Mixture-of-Experts (MoE), Generative Text Model
Use Cases 
Limitations:
Lacks alignment for instruct/chat interactions
Considerations:
Jamba should have safety moderation mechanisms and guardrails added for responsible and safe use.
Additional Notes 
The model is available for fine-tuning using techniques like PEFT.
Training Details 
Context Length:
256000
Model Architecture:
Hybrid SSM-Transformer
Input Output 
Input Format:
Compatible with 'transformers' API.
Accepted Modalities:
text
Output Format:
Generative text output
Performance Tips:
Use 'transformers' 4.40.0 or higher for optimal performance; install 'mamba-ssm' and 'causal-conv1d' for optimized Mamba implementation
LLM NameJamba V0.1
Repository 🤗https://huggingface.co/ai21labs/Jamba-v0.1 
Model Size51.6b
Required VRAM103.4 GB
Updated2025-09-23
Maintainerai21labs
Model Typejamba
Model Files  5.0 GB: 1-of-21   4.9 GB: 2-of-21   5.0 GB: 3-of-21   5.0 GB: 4-of-21   5.0 GB: 5-of-21   4.9 GB: 6-of-21   4.9 GB: 7-of-21   5.0 GB: 8-of-21   4.9 GB: 9-of-21   4.9 GB: 10-of-21   4.9 GB: 11-of-21   4.9 GB: 12-of-21   4.9 GB: 13-of-21   5.0 GB: 14-of-21   4.9 GB: 15-of-21   4.9 GB: 16-of-21   4.9 GB: 17-of-21   4.9 GB: 18-of-21   5.0 GB: 19-of-21   4.9 GB: 20-of-21   4.7 GB: 21-of-21
Model ArchitectureJambaForCausalLM
Licenseapache-2.0
Context Length262144
Model Max Length262144
Transformers Version4.40.1
Tokenizer ClassLlamaTokenizer
Padding Token<|pad|>
Vocabulary Size65536
Torch Data Typebfloat16

Quantized Models of the Jamba V0.1

Model
Likes
Downloads
VRAM
Jamba Bagel 4bit2629 GB

Best Alternatives to Jamba V0.1

Best Alternatives
Context / RAM
Downloads
Likes
Jamba V0.1 Claude Chat256K / 103.4 GB63
Jamba V0.1 Hf256K / 103.4 GB53
Code Jamba V0.10K / 103.4 GB77
Bagel Jamba V050K / 103.6 GB77
Jamba V0.1 Chat Multilingual0K / 103.4 GB2223
Airoboros Jamba 3 30K / 103.6 GB66

Rank the Jamba V0.1 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? 51544 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