LLM EXPLORER 56,604 MODELS INDEXED

Lima Unchained 70B by pankajmathur

By pankajmathur · 141 downloads
HF Repo ↗ Copy URL

Lima Unchained 70B is an open-source language model by pankajmathur. Features: 70b LLM, VRAM: 138GB, Context: 2K, License: llama2, Quantized, LLM Explorer Score: 0.1, Arc: 68.3, HellaSwag: 87.7, MMLU: 70, GSM8K: 34.7.

  Arxiv:2305.11206 Dataset:psmathur/lima unchaine...   En   Endpoints compatible   Fp16   Llama   Model-index   Pytorch   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/pankajmathur/Lima_Unchained_70b 

Lima Unchained 70B Parameters and Internals

Model Type 
text-generation
Use Cases 
Limitations:
Possibility of generating inappropriate, biased, or offensive content.
Considerations:
Exercise caution and verify information.
Training Details 
Data Sources:
psmathur/lima_unchained_v1
Methodology:
Fine-tuning using QLora on linear layers with selected ~900 conversations from the Lima dataset.
Input Output 
Input Format:
Expected format for prompts as seen in example usage.
Accepted Modalities:
text
Output Format:
Generated text responses.
Performance Tips:
Use optimal hyperparameters for generation such as top_p=0.95, top_k=0, max_new_tokens=4096.
LLM NameLima Unchained 70b
Repository 🤗https://huggingface.co/pankajmathur/Lima_Unchained_70b 
Model Size70b
Required VRAM138 GB
Updated2026-07-10
Maintainerpankajmathur
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   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 Length2048
Model Max Length2048
Transformers Version4.32.0.dev0
Tokenizer ClassLlamaTokenizer
Beginning of Sentence Token<s>
End of Sentence Token</s>
Unk Token<unk>
Vocabulary Size32000
Torch Data Typefloat16

Best Alternatives to Lima Unchained 70B

Best Alternatives
Context / RAM
Downloads
Likes
...B Instruct Gradient 1048K 8bit1024K / 75 GB415
...B Instruct Gradient 1048K 4bit1024K / 39.7 GB253
...B Instruct Gradient 1048K 2bit1024K / 21.9 GB252
...0B Instruct Gradient 262K 4bit256K / 39.7 GB253
...0B Instruct Gradient 262K 8bit256K / 75 GB262
...0B Instruct Gradient 262K 2bit256K / 21.9 GB201
... Gradient 262K 2.25bpw H6 EXL2256K / 22.2 GB80
...t Gradient 262K 4.0bpw H6 EXL2256K / 37.2 GB71
...t Gradient 262K 2.4bpw H6 EXL2256K / 23.5 GB60
...t Gradient 262K 3.5bpw H6 EXL2256K / 32.9 GB60
Note: green Score (e.g. "73.2") means that the model is better than pankajmathur/Lima_Unchained_70b.