Model Card on HF 🤗: https://huggingface.co/sigmareaver/GPT-NeoX-20b-Erebus-4bit-gptq
GPT NeoX 20B Erebus 4bit Gptq Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
GPT NeoX 20B Erebus 4bit Gptq Parameters and Internals
| LLM Name | GPT NeoX 20B Erebus 4bit Gptq |
| Repository 🤗 | https://huggingface.co/sigmareaver/GPT-NeoX-20b-Erebus-4bit-gptq |
| Model Size | 20b |
| Required VRAM | 11.2 GB |
| Updated | 2025-09-23 |
| Maintainer | sigmareaver |
| Model Type | gpt_neox |
| Model Files | |
| GPTQ Quantization | Yes |
| Quantization Type | gptq|4bit |
| Model Architecture | GPTNeoXForCausalLM |
| Context Length | 2048 |
| Model Max Length | 2048 |
| Transformers Version | 4.19.0.dev0 |
| Tokenizer Class | GPTNeoXTokenizer |
| Vocabulary Size | 50432 |
| Torch Data Type | float16 |
Best Alternatives to GPT NeoX 20B Erebus 4bit Gptq
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| GPT NeoXT Chat Base 20B | 2K / 41.2 GB | 2034 | 695 |
| EleutherAI GPT Neox 20B 4bits | 2K / 12.5 GB | 7 | 0 |
| ...t Gm Oasst1 Multilang 1024 20B | 2K / 41.2 GB | 1924 | 10 |
| H2ogpt Gm Oasst1 En 1024 20B | 2K / 41.2 GB | 1909 | 4 |
| H2ogpt Oasst1 512 20B | 2K / 41.2 GB | 1954 | 40 |
| GPT Neox 20B Full Precision | 2K / 82.5 GB | 1880 | 0 |
| Oasst GPT Neox 20B 1000 Steps | 2K / 41.2 GB | 1881 | 0 |
| Oasst GPT Neox 20B 3000 Steps | 2K / 41.2 GB | 1883 | 0 |
| GPT Neox 20B | 2K / 40.8 GB | 27209 | 573 |
| GPTNeoX 20B TestGen Dart V1.0 | 2K / 41.2 GB | 5 | 2 |
Note: green Score (e.g. "73.2") means that the model is better than sigmareaver/GPT-NeoX-20b-Erebus-4bit-gptq .
Rank the GPT NeoX 20B Erebus 4bit Gptq 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! 🌟
What open-source LLMs or SLMs are you in search of? 51534 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Release v20241124