What are the hardware requirements for Sorceroboros 33B S2a4 Gptq?

Sorceroboros 33B S2a4 Gptq requires approximately 17.6 GB of VRAM and supports a context window of 8K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Sorceroboros 33B S2a4 Gptq and how large is it?

Sorceroboros 33B S2a4 Gptq is developed by chargoddard, a model with 33b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Sorceroboros 33B S2a4 Gptq?

Sorceroboros 33B S2a4 Gptq is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Sorceroboros 33B S2a4 Gptq by chargoddard — VRAM 17.6GB, 8K context

Name: Sorceroboros 33B S2a4 Gptq
Author: chargoddard

Sorceroboros 33B S2a4 Gptq is an open-source language model by chargoddard. Features: 33b LLM, VRAM: 17.6GB, Context: 8K, Quantized, Instruction-Based, Merged, LLM Explorer Score: 0.08.

Merged Model 4bit Custom code Dataset:ehartford/wizard vicun... Dataset:ehartford/wizardlm evo... Dataset:jondurbin/airoboros-gp... Dataset:openai/summarize from ... En Endpoints compatible Gptq Instruct Llama Quantized Region:us

Model Card on HF 🤗: https://huggingface.co/chargoddard/sorceroboros-33b-s2a4-gptq

Sorceroboros 33B S2a4 Gptq Parameters and Internals

Model Type

text generation

Additional Notes

Ensure both `compress_pos_emb` (or `scale`) is set to 2, and `alpha_value` is set to 4 when loading.

Supported Languages

en (English)

Training Details

Data Sources:

ehartford/WizardLM_evol_instruct_V2_196k_unfiltered_merged_split, jondurbin/airoboros-gpt4-1.4.1, openai/summarize_from_feedback, ehartford/wizard_vicuna_70k_unfiltered

Methodology:

This model was trained using both linear and NTK-aware RoPE scaling in tandem.

Context Length:

8192

Input Output

Input Format:

vicuna 1.1

LLM Name	Sorceroboros 33B S2a4 Gptq
Repository 🤗	https://huggingface.co/chargoddard/sorceroboros-33b-s2a4-gptq
Merged Model	Yes
Model Size	33b
Required VRAM	17.6 GB
Updated	2026-08-06
Maintainer	chargoddard
Model Type	llama
Instruction-Based	Yes
Model Files	17.6 GB
Supported Languages	en
GPTQ Quantization	Yes
Quantization Type	gptq\|4bit
Model Architecture	LlamaForCausalLM
Context Length	8192
Model Max Length	8192
Transformers Version	4.30.0.dev0
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
Vocabulary Size	32000
Torch Data Type	float16

Rank the Sorceroboros 33B S2a4 Gptq capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

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

Best Alternatives to Sorceroboros 33B S2a4 Gptq

Best Alternatives	Context / RAM	Downloads	Likes
...epseek Coder 33B Instruct GPTQ	16K / 17.4 GB	267	25
Med Orca Instruct 33B GPTQ	2K / 17.6 GB	10	1
...er 33B Instruct 4.0bpw H6 EXL2	16K / 17.1 GB	5	5
...er 33B Instruct 8.0bpw H8 EXL2	16K / 33.6 GB	6	3
...er 33B Instruct 3.0bpw H6 EXL2	16K / 13 GB	10	1
...r 33B Instruct 4.65bpw H6 EXL2	16K / 19.8 GB	6	1
...er 33B Instruct 5.0bpw H6 EXL2	16K / 21.2 GB	5	1
...ardLM 33B V1.0 Uncensored GPTQ	2K / 16.9 GB	47	42
Deepseek Wizard 33B Slerp	16K / 35.3 GB	7	0
ValidateAI 3 33B Ties	16K / 66.5 GB	7	0

Note: green Score (e.g. "73.2") means that the model is better than chargoddard/sorceroboros-33b-s2a4-gptq.