LLM EXPLORER 59,420 MODELS INDEXED

GPT J 6B Skein 4bit 128g by Tokenisrequired

By Tokenisrequired · 13 downloads
HF Repo ↗ Copy URL

GPT J 6B Skein 4bit 128g is an open-source language model by Tokenisrequired. Features: 6b LLM, VRAM: 4GB, Quantized, LLM Explorer Score: 0.06.

  4bit   Endpoints compatible   Gptj   Quantized   Region:us Text-generation 4bit int4 kobo...
Model Card on HF 🤗: https://huggingface.co/Tokenisrequired/GPT-J-6B-Skein-4bit-128g 

GPT J 6B Skein 4bit 128g Parameters and Internals

Model Type 
text-generation
Additional Notes 
The model outputs special characters even if the context does not have any, and occasionally transforms names inaccurately. It can generate incorrect pronouns, split actions unnaturally, and fails to produce dialogue tags. The model encountered a 'LayerNormKernelImpl' not implemented for 'Half' error. It's optimized for GTX 1080 with 4bit inference through GPTQ, loaded with specific settings via Python script.
LLM NameGPT J 6B Skein 4bit 128g
Repository 🤗https://huggingface.co/Tokenisrequired/GPT-J-6B-Skein-4bit-128g 
Model Size6b
Required VRAM4 GB
Updated2026-08-06
MaintainerTokenisrequired
Model Typegptj
Model Files  4.0 GB
Quantization Type4bit
Model ArchitectureGPTJForCausalLM
Model Max Length2048
Transformers Version4.10.0.dev0
Tokenizer ClassGPT2Tokenizer
Vocabulary Size50400
Torch Data Typefloat16
Activation Functiongelu_new

Best Alternatives to GPT J 6B Skein 4bit 128g

Best Alternatives
Context / RAM
Downloads
Likes
Model0K / 6.2 GB110
...oduct NER GPT J 6B 4bit Merged0K / 2.5 GB120
...nese Novel GPT J 6B F16 Marisa0K / 12.2 GB43
Kakaobrain Kogpt 6B 8bit0K / 6.7 GB52
Pygmalion 6b Dev 4bit 128g0K / 4 GB8121
GPT J 6B Alpaca Gpt40K / 24.3 GB4620
Pygmalion 6B 4bit 128g0K / 4 GB133
Dolly Shygmalion 6B 4bit 128g0K / 4 GB145
Pygmalion 6B 4bit 128g0K / 4 GB1740
...O Pygway V8p4 Dev 6B 4bit 128g0K / 4 GB82
Note: green Score (e.g. "73.2") means that the model is better than Tokenisrequired/GPT-J-6B-Skein-4bit-128g.