What are the hardware requirements for Hamanasu KTO V2?

Hamanasu KTO V2 requires approximately 9 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 Hamanasu KTO V2 and how large is it?

Hamanasu KTO V2 is developed by NewEden, a model with 4.5b 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 Hamanasu KTO V2?

Hamanasu KTO V2 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Hamanasu KTO V2 by NewEden — VRAM 9GB, 128K context

Name: Hamanasu KTO V2
Author: NewEden

Hamanasu KTO V2 is an open-source language model by NewEden. Features: 4.5b LLM, VRAM: 9GB, Context: 128K, LLM Explorer Score: 0.17.

Llama Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/NewEden/Hamanasu-KTO-V2

Hamanasu KTO V2 Parameters and Internals

LLM Name	Hamanasu KTO V2
Repository 🤗	https://huggingface.co/NewEden/Hamanasu-KTO-V2
Model Size	4.5b
Required VRAM	9 GB
Updated	2025-04-06
Maintainer	NewEden
Model Type	llama
Model Files	5.0 GB: 1-of-2 4.0 GB: 2-of-2 0.0 GB 0.0 GB
Model Architecture	LlamaForCausalLM
Context Length	131072
Model Max Length	131072
Transformers Version	4.50.0
Tokenizer Class	PreTrainedTokenizer
Padding Token	<\|finetune_right_pad_id\|>
Vocabulary Size	128256
Torch Data Type	bfloat16

Rank the Hamanasu KTO V2 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 Hamanasu KTO V2

Best Alternatives	Context / RAM	Downloads	Likes
Mergefodder2	128K / 9.1 GB	7	0
Mergefodder1	128K / 9.1 GB	7	0
Basal 1.0 4.5B	32K / 9.5 GB	1743	7
Basal 1.5 4.5B	32K / 9.5 GB	470	7
...ik 4.5B V3 Matura History Sft2	32K / 9.5 GB	636	1

Note: green Score (e.g. "73.2") means that the model is better than NewEden/Hamanasu-KTO-V2.