LLM EXPLORER 59,358 MODELS INDEXED

DeepSeek R1 MFANN TIES Unretrained 7B by netcat420

By netcat420 · 34 downloads
HF Repo ↗ Copy URL

DeepSeek R1 MFANN TIES Unretrained 7B is an open-source language model by netcat420. Features: 7b LLM, VRAM: 15.2GB, Context: 4K, Code Generating, Merged, LLM Explorer Score: 0.18.

  Merged Model   Arxiv:2306.01708   Autotrain compatible Base model:deepseek-ai/deepsee... Base model:netcat420/qwen2.5-c... Base model:qwen/qwen2.5-math-7...   Codegen   Conversational   Endpoints compatible   Qwen2   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/netcat420/DeepSeek-R1-MFANN-TIES-unretrained-7b 

DeepSeek R1 MFANN TIES Unretrained 7B Parameters and Internals

LLM NameDeepSeek R1 MFANN TIES Unretrained 7B
Repository 🤗https://huggingface.co/netcat420/DeepSeek-R1-MFANN-TIES-unretrained-7b 
Base Model(s)  Qwen/Qwen2.5-Math-7B   DeepSeek R1 Distill Qwen 7B   netcat420/Qwen2.5-Coder-Scholar-7B-Abliterated-MFANN-Slerp-Unretrained   Qwen/Qwen2.5-Math-7B   deepseek-ai/DeepSeek-R1-Distill-Qwen-7B   netcat420/Qwen2.5-Coder-Scholar-7B-Abliterated-MFANN-Slerp-Unretrained
Merged ModelYes
Model Size7b
Required VRAM15.2 GB
Updated2025-01-23
Maintainernetcat420
Model Typeqwen2
Model Files  5.0 GB: 1-of-4   4.9 GB: 2-of-4   5.0 GB: 3-of-4   0.3 GB: 4-of-4
Generates CodeYes
Model ArchitectureQwen2ForCausalLM
Context Length4096
Model Max Length4096
Transformers Version4.46.2
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size152064
Torch Data Typefloat16
Errorsreplace

Best Alternatives to DeepSeek R1 MFANN TIES Unretrained 7B

Best Alternatives
Context / RAM
Downloads
Likes
SakalFusion 7B Beta986K / 15.2 GB90
Qwen2.5 7B Rebase986K / 15.2 GB732
Qwen2.5 7B Rebase986K / 15.2 GB162
...R1 Distill Qwen MFANN Slerp 7B128K / 15.2 GB110
Qwen2.5 7B Coder Codeio Pp128K / 15.2 GB85
Qwen2.5 7B CySecButler V0.1128K / 15.2 GB93
CoT 2.5128K / 15.2 GB390
Mergekit Ties Uqhfast128K / 15.2 GB260
CoT 2.5128K / 15.2 GB260
StockQwen 2.5 7B128K / 15.2 GB696