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

Synthia 70B V1.2B by migtissera

 »  All LLMs  »  migtissera  »  Synthia 70B V1.2B   URL Share it on

  819   

Synthia 70B V1.2B is an open-source language model by migtissera. Features: 70b LLM, VRAM: 138GB, Context: 4K, License: llama2, Quantized, HF Score: 67, LLM Explorer Score: 0.12, Arc: 68.8, HellaSwag: 87.6, MMLU: 68.8, TruthfulQA: 57.7, WinoGrande: 83.9, GSM8K: 35.3.

  Arxiv:2302.13971   Arxiv:2306.02707   Deploy:azure   En   Endpoints compatible   Fp16   Llama   Pytorch   Quantized   Region:us   Sharded
Model Card on HF 🤗: https://huggingface.co/migtissera/Synthia-70B-v1.2b 

Synthia 70B V1.2B Benchmarks

ARC: 68.77 vs 96.7 (so35)-28.9%
HellaSwag: 87.57 vs 95.3 (gpt4)-8.1%
MMLU: 68.81 vs 88.3 (so35)-22.1%
TruthfulQA: 57.69 vs 59 (gpt4)-2.2%
WinoGrande: 83.9 vs 87.5 (gpt4)-4.1%
GSM8K: 35.25 vs 96.4 (so35)-63.4%
LLME Score: 0.11792
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Synthia 70B V1.2B (migtissera/Synthia-70B-v1.2b)
🌟 Advertise your project 🚀

Synthia 70B V1.2B Parameters and Internals

Model Type 
LLM, text generation
Additional Notes 
While the model aims for accuracy, it can produce inaccurate or misleading results. It may generate inappropriate, biased, or offensive content.
Supported Languages 
en (English)
Training Details 
Data Sources:
Orca style datasets
Methodology:
fine-tuned for instruction following and long-form conversation
Training Time:
14 days for 1 epoch
Model Architecture:
LLama-2-70B
Input Output 
Input Format:
Prompt format with 'SYSTEM,' 'USER,' and 'ASSISTANT' labels
Accepted Modalities:
text
Output Format:
Text
Performance Tips:
Use Tree of Thoughts reasoning for coherent answers.
LLM NameSynthia 70B V1.2B
Repository 🤗https://huggingface.co/migtissera/Synthia-70B-v1.2b 
Base Model(s)  migtissera/Synthia-70B-v1.2   migtissera/Synthia-70B-v1.2
Model Size70b
Required VRAM138 GB
Updated2026-02-16
Maintainermigtissera
Model Typellama
Model Files  9.8 GB: 1-of-15   9.8 GB: 2-of-15   10.0 GB: 3-of-15   9.8 GB: 4-of-15   9.8 GB: 5-of-15   9.8 GB: 6-of-15   10.0 GB: 7-of-15   9.8 GB: 8-of-15   9.8 GB: 9-of-15   9.8 GB: 10-of-15   10.0 GB: 11-of-15   9.8 GB: 12-of-15   9.8 GB: 13-of-15   9.5 GB: 14-of-15   0.5 GB: 15-of-15
Supported Languagesen
Quantization Typefp16
Model ArchitectureLlamaForCausalLM
Licensellama2
Context Length4096
Model Max Length4096
Transformers Version4.31.0
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size32000
Torch Data Typefloat16

Quantized Models of the Synthia 70B V1.2B

Model
Likes
Downloads
VRAM
Synthia 70B V1.2B GGUF2024429 GB
Synthia 70B V1.2B AWQ1736 GB
Synthia 70B V1.2B GPTQ8035 GB

Best Alternatives to Synthia 70B V1.2B

Best Alternatives
Context / RAM
Downloads
Likes
...B Instruct Gradient 1048K 8bit1024K / 75 GB445
...B Instruct Gradient 1048K 4bit1024K / 39.7 GB123
...B Instruct Gradient 1048K 2bit1024K / 21.9 GB172
...0B Instruct Gradient 262K 4bit256K / 39.7 GB183
...0B Instruct Gradient 262K 8bit256K / 75 GB52
...0B Instruct Gradient 262K 2bit256K / 21.9 GB91
... Gradient 262K 2.25bpw H6 EXL2256K / 22.2 GB80
...t Gradient 262K 4.0bpw H6 EXL2256K / 37.2 GB41
...t Gradient 262K 3.5bpw H6 EXL2256K / 32.9 GB60
...t Gradient 262K 2.4bpw H6 EXL2256K / 23.5 GB50
Note: green Score (e.g. "73.2") means that the model is better than migtissera/Synthia-70B-v1.2b.

Rank the Synthia 70B V1.2B 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? 52392 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