What are the hardware requirements for Cassandra 6.9B GPTQ?

Cassandra 6.9B GPTQ requires approximately 4.7 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Cassandra 6.9B GPTQ and how large is it?

Cassandra 6.9B GPTQ is developed by TheBloke, a model with 6.9b 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 Cassandra 6.9B GPTQ?

Cassandra 6.9B GPTQ is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Cassandra 6.9B GPTQ by TheBloke — VRAM 4.7GB, 4K context

Name: Cassandra 6.9B GPTQ
Author: TheBloke

Cassandra 6.9B GPTQ is an open-source language model by TheBloke. Features: 6.9b LLM, VRAM: 4.7GB, Context: 4K, License: other, Quantized, Fine-Tuned, LLM Explorer Score: 0.07.

4-bit Finetuned Gpt neox Gptq Quantized Region:us Safetensors

Model Card on HF 🤗: https://huggingface.co/TheBloke/cassandra-6.9B-GPTQ

Cassandra 6.9B GPTQ Parameters and Internals

Model Type

text generation

Additional Notes

Quantized to 4bit using AutoGPTQ for optimized inference accuracy and speed. Supports varying bit sizes for CPU+GPU inference.

Input Output

Input Format:

NovelAI-style

Accepted Modalities:

text

Performance Tips:

Use '>" as a stop token in your UI

LLM Name	Cassandra 6.9B GPTQ
Repository 🤗	https://huggingface.co/TheBloke/cassandra-6.9B-GPTQ
Model Size	6.9b
Required VRAM	4.7 GB
Updated	2026-08-06
Maintainer	TheBloke
Model Type	gpt_neox
Model Files	4.7 GB
GPTQ Quantization	Yes
Quantization Type	gptq
Model Architecture	GPTNeoXForCausalLM
License	other
Context Length	4096
Model Max Length	4096
Transformers Version	4.27.2
Vocabulary Size	50277
Torch Data Type	float16

Rank the Cassandra 6.9B 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 Cassandra 6.9B GPTQ

Best Alternatives	Context / RAM	Downloads
...Int4 Step107000 GPTQ Wikitext2	2K / 4.2 GB	7
...Int3 Step107000 GPTQ Wikitext2	2K / 3.4 GB	6
... Int3 Step36000 GPTQ Wikitext2	2K / 3.4 GB	5
... Int3 Step71000 GPTQ Wikitext2	2K / 3.4 GB	5
...Int3 Step143000 GPTQ Wikitext2	2K / 3.4 GB	5
...Int3 Step110000 GPTQ Wikitext2	2K / 3.4 GB	5
...Int4 Step110000 GPTQ Wikitext2	2K / 4.2 GB	6
... Int4 Step71000 GPTQ Wikitext2	2K / 4.2 GB	5
...Int4 Step143000 GPTQ Wikitext2	2K / 4.2 GB	5
...ythia 6.9B Int4 GPTQ Wikitext2	2K / 4.2 GB	7

Note: green Score (e.g. "73.2") means that the model is better than TheBloke/cassandra-6.9B-GPTQ.