What are the hardware requirements for Lev1 3.7B?

Lev1 3.7B requires approximately 10.2 GB of VRAM and supports a context window of 512K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Lev1 3.7B and how large is it?

Lev1 3.7B is developed by theaviv, a model with 3.7b 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 Lev1 3.7B?

Lev1 3.7B is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Lev1 3.7B by theaviv — VRAM 10.2GB, 512K context

Name: Lev1 3.7B
Author: theaviv

Lev1 3.7B is an open-source language model by theaviv. Features: 3.7b LLM, VRAM: 10.2GB, Context: 512K, License: apache-2.0, LLM Explorer Score: 0.32.

Base model:adapter:ifm/k2-hori... Base model:ifm/k2-horizon-3.7b Classification Conversational Custom code Dataset:openai/gsm8k Dataset:theaviv/lev1-decisions Decision-index Decision-model En K2-horizon K2 horizon Lora Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/theaviv/Lev1-3.7B

Lev1 3.7B Parameters and Internals

LLM Name	Lev1 3.7B
Repository 🤗	https://huggingface.co/theaviv/Lev1-3.7B
Base Model(s)	IFM/K2-Horizon-3.7B IFM/K2-Horizon-3.7B
Model Size	3.7b
Required VRAM	10.2 GB
Updated	2026-10-08
Maintainer	theaviv
Model Type	k2_horizon
Model Files	1.3 GB: 1-of-10 1.3 GB: 2-of-10 1.0 GB: 3-of-10 1.0 GB: 4-of-10 1.0 GB: 5-of-10 1.0 GB: 6-of-10 1.0 GB: 7-of-10 1.0 GB: 8-of-10 1.0 GB: 9-of-10 0.6 GB: 10-of-10
Supported Languages	en
Model Architecture	K2HorizonForCausalLM
License	apache-2.0
Context Length	524288
Model Max Length	524288
Transformers Version	5.17.0
Tokenizer Class	TokenizersBackend
Padding Token	<\|ifm\|endoftext\|>
Vocabulary Size	250624

Rank the Lev1 3.7B 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 Lev1 3.7B

Best Alternatives	Context / RAM	Downloads	Likes
K2 Horizon 3.7B	512K / 9.8 GB	1765	17
Ocean 3.7B Agentic	512K / 9.8 GB	300	1
K2 Horizon 3.7B Heretic	512K / 10.2 GB	132	0
NicomEdu Alpha	512K / 10.2 GB	316	1
K2 Horizon 3.7B MLX 8bit	512K / 5.3 GB	229	2