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

Codallama 7B Instruct Nf4 Fp16 Upscaled by arnavgrg

 »  All LLMs  »  arnavgrg  »  Codallama 7B Instruct Nf4 Fp16 Upscaled   URL Share it on

  15   
  Autotrain compatible   Codegen   Endpoints compatible   Fp16   Instruct   Llama   Quantized   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/arnavgrg/codallama-7b-instruct-nf4-fp16-upscaled 

Codallama 7B Instruct Nf4 Fp16 Upscaled Benchmarks

LLME Score: 0.15741
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
Codallama 7B Instruct Nf4 Fp16 Upscaled (arnavgrg/codallama-7b-instruct-nf4-fp16-upscaled)

Codallama 7B Instruct Nf4 Fp16 Upscaled Parameters and Internals

Model Type 
text generation, inference
Additional Notes 
Quantization operation to nf4 is not lossless; model weights for linear layers are lossy
Training Details 
Methodology:
Upscaled fp16 variant with nf4 4-bit quantization
Model Architecture:
Linear4bit layers upscaled to fp16
Input Output 
Accepted Modalities:
text
Performance Tips:
Upscaling linear4bit layers to fp16 reduces overhead from quantization/dequantization
LLM NameCodallama 7B Instruct Nf4 Fp16 Upscaled
Repository 🤗https://huggingface.co/arnavgrg/codallama-7b-instruct-nf4-fp16-upscaled 
Model Size7b
Required VRAM13.5 GB
Updated2024-12-22
Maintainerarnavgrg
Model Typellama
Instruction-BasedYes
Model Files  4.9 GB: 1-of-3   5.0 GB: 2-of-3   3.6 GB: 3-of-3
Quantization Typefp16
Generates CodeYes
Model ArchitectureLlamaForCausalLM
Licenseapache-2.0
Context Length16384
Model Max Length16384
Transformers Version4.35.2
Tokenizer ClassCodeLlamaTokenizer
Padding Token[PAD]
Vocabulary Size32016
Torch Data Typefloat16

Best Alternatives to Codallama 7B Instruct Nf4 Fp16 Upscaled

Best Alternatives
Context / RAM
Downloads
Likes
...ruct Solidity Bnb 4bit Smashed16K / 4.2 GB140
...B Instruct Hf Bnb 4bit Smashed16K / 4.2 GB210
CodelLama7B Inst DPO 7K Mlx16K / 4.2 GB82
...eLlama 7B Instruct Hf 4bit MLX16K / 4.2 GB121
...6.7B Instruct 8.0bpw H8 EXL2 216K / 6.8 GB92
...6.7B Instruct 3.0bpw H6 EXL2 216K / 2.8 GB91
CodeLlama 7B Instruct Fp1616K / 13.5 GB338
...Llama 7B Instruct Bf16 Sharded16K / 13.5 GB161
...B Instruct V1.5 6.0bpw H6 EXL24K / 5.7 GB72
...B Instruct V1.5 8.0bpw H8 EXL24K / 7.3 GB111
Note: green Score (e.g. "73.2") means that the model is better than arnavgrg/codallama-7b-instruct-nf4-fp16-upscaled.

Rank the Codallama 7B Instruct Nf4 Fp16 Upscaled 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? 40066 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 v20241217
  Support LLM Explorer