KNK VF Lab 38B is an open-source language model by MagistrTheOne. Features: 38b LLM, VRAM: 76.2GB, Context: 128K, License: other, LLM Explorer Score: 0.28.
B300 Bf16 Code En Endpoints compatible Enterprise H200 Initialization Knk-vf Knk vf Knkf Kurotama-no-kami Long-context Megatron Mixture-of-experts Moe Multilingual Nullxes Random-init Region:us Ru Safetensors Sharded Sparse-moe Tensorflow Void-forged
Model Card on HF 🤗: https://huggingface.co/MagistrTheOne/KNK-VF-Lab-38B
KNK VF Lab 38B Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
KNK VF Lab 38B Parameters and Internals
| LLM Name | KNK VF Lab 38B |
| Repository 🤗 | https://huggingface.co/MagistrTheOne/KNK-VF-Lab-38B |
| Model Creator | NULLXES |
| Model Size | 38b |
| Required VRAM | 76.2 GB |
| Updated | 2026-06-08 |
| Maintainer | MagistrTheOne |
| Model Type | knk_vf |
| Model Files | |
| Supported Languages | en ru |
| Model Architecture | KNKVFForCausalLM |
| License | other |
| Context Length | 131072 |
| Model Max Length | 131072 |
| Vocabulary Size | 128000 |
| Torch Data Type | bfloat16 |
Rank the KNK VF Lab 38B Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 54413 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a