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

Deepseek Coder 6.7B Instruct AWQ by TheBloke

 »  All LLMs  »  TheBloke  »  Deepseek Coder 6.7B Instruct AWQ   URL Share it on

  1736   
  4-bit   Autotrain compatible   Awq Base model:deepseek-ai/deepsee... Base model:quantized:deepseek-...   Codegen   Conversational   Instruct   Llama   Quantized   Region:us   Safetensors

Deepseek Coder 6.7B Instruct AWQ Benchmarks

LLME Score: 0.17678
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

Deepseek Coder 6.7B Instruct AWQ Parameters and Internals

Model Type 
deepseek
Additional Notes 
TheBloke has quantized the model using AWQ methodology, supporting efficient, fast low-bit weight quantization, currently supporting 4-bit quantization.
Supported Languages 
English (87%), Chinese (13%)
Training Details 
Data Sources:
2T tokens of code and linguistic data in both English and Chinese languages
Data Volume:
2T tokens
Methodology:
window size of 16K and an extra fill-in-the-blank task
Context Length:
16384
Input Output 
Input Format:
"You are an AI programming assistant, utilizing the Deepseek Coder model, developed by Deepseek Company, and you only answer questions related to computer science. For politically sensitive questions, security and privacy issues, and other non-computer science questions, you will refuse to answer. ### Instruction: {prompt} ### Response: "
Accepted Modalities:
text
Output Format:
text
LLM NameDeepseek Coder 6.7B Instruct AWQ
Repository 🤗https://huggingface.co/TheBloke/deepseek-coder-6.7B-instruct-AWQ 
Model NameDeepseek Coder 6.7B Instruct
Model CreatorDeepSeek
Base Model(s)  deepseek-ai/deepseek-coder-6.7b-instruct   deepseek-ai/deepseek-coder-6.7b-instruct
Model Size6.7b
Required VRAM3.9 GB
Updated2024-11-21
MaintainerTheBloke
Model Typedeepseek
Instruction-BasedYes
Model Files  3.9 GB
AWQ QuantizationYes
Quantization Typeawq
Generates CodeYes
Model ArchitectureLlamaForCausalLM
Licenseother
Context Length16384
Model Max Length16384
Transformers Version4.35.0
Tokenizer ClassLlamaTokenizerFast
Beginning of Sentence Token<｜begin▁of▁sentence｜>
End of Sentence Token<|EOT|>
Vocabulary Size32256
Torch Data Typefloat16
Deepseek Coder 6.7B Instruct AWQ (TheBloke/deepseek-coder-6.7B-instruct-AWQ)

Best Alternatives to Deepseek Coder 6.7B Instruct AWQ

Best Alternatives
Context / RAM
Downloads
Likes
...rpreter DS 6.7B 6.0bpw H6 EXL216K / 5.2 GB22
...rpreter DS 6.7B 8.0bpw H8 EXL216K / 6.9 GB22
...rpreter DS 6.7B 4.0bpw H6 EXL216K / 3.6 GB21
...s Coder6.7b Reflct Adamw Iter116K / 13.5 GB3000
Speechless Coder Ds 6.7B16K / 13.5 GB12685
Code 290K 6.7B Instruct16K / 13.5 GB886
Datascience Coder 6.7B16K / 13.5 GB12453
Hpc Coder V2.6.7B16K / 27.1 GB2226
Merge Code Gen16K / 13.6 GB140
... 6.7B Instruct Trt Int4 G64 Hf16K / 4.2 GB60
Note: green Score (e.g. "73.2") means that the model is better than TheBloke/deepseek-coder-6.7B-instruct-AWQ.

Rank the Deepseek Coder 6.7B Instruct AWQ 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? 38199 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 v20241110
  Support LLM Explorer