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

SAM by SuperAGI

 »  All LLMs  »  SuperAGI  »  SAM   URL Share it on

  710   

SAM is an open-source language model by SuperAGI. Features: 7b LLM, VRAM: 14.4GB, Context: 32K, License: apache-2.0, HF Score: 59.3, LLM Explorer Score: 0.14, Arc: 59.4, HellaSwag: 82.3, MMLU: 62.2, TruthfulQA: 52.6, WinoGrande: 76.4, GSM8K: 22.9.

  En   Endpoints compatible   Mistral   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/SuperAGI/SAM 

SAM Benchmarks

ARC: 59.39 vs 96.7 (so35)-38.6%
HellaSwag: 82.31 vs 95.3 (gpt4)-13.6%
MMLU: 62.15 vs 88.3 (so35)-29.6%
TruthfulQA: 52.64 vs 59 (gpt4)-10.8%
WinoGrande: 76.4 vs 87.5 (gpt4)-12.7%
GSM8K: 22.9 vs 96.4 (so35)-76.2%
LLME Score: 0.14268
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
SAM (SuperAGI/SAM)
🌟 Advertise your project 🚀

SAM Parameters and Internals

Model Type 
reasoning
Use Cases 
Areas:
reasoning, task breakdown
Limitations:
Not suitable for conversations and simple Q&A, Lacks moderation mechanisms for toxicity and societal bias
Considerations:
Not suitable for production usage due to lack of guardrails
Additional Notes 
Demonstrates the potential for high-quality data with less volume to induce better reasoning.
Supported Languages 
en (primary)
Training Details 
Data Volume:
97% smaller dataset compared to others
Methodology:
Fine-tuned using responses generated by open-source models
Training Time:
4 hours
Hardware Used:
NVIDIA 6 x H100 SxM (80GB)
Model Architecture:
Trained using Mistral 7B architecture
Input Output 
Input Format:
Template with [INST] and [/INST] for instructions
Accepted Modalities:
text
Output Format:
Textual generation
Performance Tips:
Suggested temperature = 0.3 for optimal performance
LLM NameSAM
Repository 🤗https://huggingface.co/SuperAGI/SAM 
Model Size7b
Required VRAM14.4 GB
Updated2026-04-03
MaintainerSuperAGI
Model Typemistral
Model Files  4.9 GB: 1-of-3   5.0 GB: 2-of-3   4.5 GB: 3-of-3
Supported Languagesen
Model ArchitectureMistralForCausalLM
Licenseapache-2.0
Context Length32768
Model Max Length32768
Transformers Version4.36.2
Tokenizer ClassLlamaTokenizer
Vocabulary Size32000
Torch Data Typefloat16

Quantized Models of the SAM

Model
Likes
Downloads
VRAM
SAM GGUF61283 GB
SAM GPTQ074 GB
SAM AWQ064 GB

Best Alternatives to SAM

Best Alternatives
Context / RAM
Downloads
Likes
...Nemo Instruct 2407 Abliterated1000K / 24.5 GB25420
MegaBeam Mistral 7B 512K512K / 14.4 GB844453
SpydazWeb AI HumanAI RP512K / 14.4 GB141
SpydazWeb AI HumanAI 002512K / 14.4 GB181
...daz Web AI ChatML 512K Project512K / 14.5 GB120
MegaBeam Mistral 7B 300K282K / 14.4 GB377916
MegaBeam Mistral 7B 300K282K / 14.4 GB808216
Hebrew Mistral 7B 200K256K / 30 GB125115
Astral 256K 7B V2250K / 14.4 GB50
Astral 256K 7B250K / 14.4 GB50
Note: green Score (e.g. "73.2") means that the model is better than SuperAGI/SAM.

Rank the SAM 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? 52509 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.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a
  Support LLM Explorer