What are the hardware requirements for SeQwence 14Bv3?

SeQwence 14Bv3 requires approximately 29.5 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed SeQwence 14Bv3 and how large is it?

SeQwence 14Bv3 is developed by CultriX, a model with 14.8b 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 SeQwence 14Bv3?

SeQwence 14Bv3 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

SeQwence 14Bv3 by CultriX — VRAM 29.5GB, 128K context

Name: SeQwence 14Bv3
Author: CultriX

SeQwence 14Bv3 is an open-source language model by CultriX. Features: 14.8b LLM, VRAM: 29.5GB, Context: 128K, Merged, LLM Explorer Score: 0.26.

Merged Model Arxiv:2306.01708 Arxiv:2311.03099 Base model:cultrix/seqwence-14... Base model:finetune:cultrix/se... Conversational Endpoints compatible Model-index Qwen2 Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/CultriX/SeQwence-14Bv3

SeQwence 14Bv3 Parameters and Internals

LLM Name	SeQwence 14Bv3
Repository 🤗	https://huggingface.co/CultriX/SeQwence-14Bv3
Base Model(s)	SeQwence 14Bv2 Qwestion 14B SeQwence 14Bv1 CultriX/SeQwence-14Bv2 CultriX/Qwestion-14B CultriX/SeQwence-14Bv1
Merged Model	Yes
Model Size	14.8b
Required VRAM	29.5 GB
Updated	2026-08-24
Maintainer	CultriX
Model Type	qwen2
Model Files	29.5 GB: 1-of-1
Model Architecture	Qwen2ForCausalLM
Context Length	131072
Model Max Length	131072
Transformers Version	4.46.3
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151665
Torch Data Type	bfloat16
Errors	replace

Rank the SeQwence 14Bv3 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 SeQwence 14Bv3

Best Alternatives	Context / RAM	Downloads	Likes
Viper Coder V1.7 Vsm6	986K / 29.7 GB	9	6
Cheng 2 Ingredient4	986K / 29.7 GB	6	0
...acilla Vipercoderv1.7vsm6 0308	986K / 29.7 GB	12	0
Temp Model For 2 Mini 006	986K / 29.7 GB	8	0
Tempmotacilla Cinerea 0308	128K / 29.7 GB	19	2
...irtuoso Small V2 Tensopolis V1	128K / 29.7 GB	5	1
Merge Model 20250308 2	128K / 29.7 GB	11	0
Tempesthenno Kto 0205 Ckpt80	128K / 29.6 GB	5	3
Viper Coder Hybrid V1.3	128K / 29.7 GB	16	9
Miisce One	128K / 29.7 GB	8	2

Note: green Score (e.g. "73.2") means that the model is better than CultriX/SeQwence-14Bv3.