What are the hardware requirements for NeoHorse 1 4B FP8?

NeoHorse 1 4B FP8 requires approximately 4.8 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed NeoHorse 1 4B FP8 and how large is it?

NeoHorse 1 4B FP8 is developed by Codeflux, a model with 4b 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 NeoHorse 1 4B FP8?

NeoHorse 1 4B FP8 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

NeoHorse 1 4B FP8 by Codeflux — VRAM 4.8GB, 256K context

Name: NeoHorse 1 4B FP8
Author: Codeflux

NeoHorse 1 4B FP8 is an open-source language model by Codeflux. Features: 4b LLM, VRAM: 4.8GB, Context: 256K, License: apache-2.0, Instruction-Based, LLM Explorer Score: 0.26.

Arxiv:2609.08183 Agentic Base model:quantized:tokenrhyt... Base model:tokenrhythm/neohors... Coding Conversational Fp8 Instruct Instruction-following Modelopt Quantized Qwen3 5 text Reasoning Region:us Safetensors Tool-use

Model Card on HF 🤗: https://huggingface.co/Codeflux/NeoHorse-1-4B-FP8

NeoHorse 1 4B FP8 Parameters and Internals

LLM Name	NeoHorse 1 4B FP8
Repository 🤗	https://huggingface.co/Codeflux/NeoHorse-1-4B-FP8
Base Model(s)	NeoHorse 1 4B TokenRhythm/NeoHorse-1-4B
Model Size	4b
Required VRAM	4.8 GB
Updated	2026-10-21
Maintainer	Codeflux
Model Type	qwen3_5_text
Instruction-Based	Yes
Model Files	4.8 GB
Model Architecture	Qwen3_5ForCausalLM
License	apache-2.0
Context Length	262144
Model Max Length	262144
Transformers Version	5.14.1
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	248320
Errors	replace

Rank the NeoHorse 1 4B FP8 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 NeoHorse 1 4B FP8

Best Alternatives	Context / RAM	Downloads	Likes
NeoHorse 1 4B	256K / 8.4 GB	1771	179
Cisimcik 4B	256K / 8.4 GB	848	1
...ihui NeoHorse 1 4B Abliterated	256K / 8.4 GB	546	4
NeoHorse 1 4B	256K / 8.4 GB	125	1
Qwen3.5 4B TR Finance	256K / 8.4 GB	26	0
NeoHorse 1 4B 4bit	256K / 2.4 GB	118	2
Mind Mem 4B	256K / 2.7 GB	22	1

Note: green Score (e.g. "73.2") means that the model is better than Codeflux/NeoHorse-1-4B-FP8.