Small language models, built from scratch.

KaiLu is the KvcHub Labs family of chat models. Every checkpoint is trained from random initialization and sized to run offline on your own machine. KaiLu 3 Pro is the newest: KaiLu 3 grown to 1B parameters and trained further, open source under Apache 2.0, and runnable from a single 1.1 GB file.

ollama run hf.co/KvcHub/KaiLu-3-Pro
Chat with it in your browser Compare the models
Smarts
  1. KaiLu 343.8 HellaSwag 36.7% · WinoGrande 50.9%
  2. KaiLu 3 Pronewest43.7 HellaSwag 35.4% · WinoGrande 52.0%
  3. KaiLu 2 2B43.3 HellaSwag 34.8% · WinoGrande 51.9%
  4. KaiLu 2 933M43.0 HellaSwag 35.0% · WinoGrande 51.0%

Average accuracy on HellaSwag (first 2,000 validation questions) and WinoGrande (all 1,267), measured by KvcHub with llama.cpp on each model's default download. Bars start at random guessing (37.5); gaps under about 1 point are within measurement noise.

Talk to KaiLu, right here

KaiLu runs inside your browser. The model downloads once and stays cached; after that, everything happens on your device and nothing you type is sent anywhere.

Not loaded

Load KaiLu 3 to start chatting.

Downloads once, then loads from your browser's cache. Best in Chrome, Edge, Firefox or Safari on a laptop or desktop.

Speed
–
Last reply
–
Runs on
–
Model
–

KaiLu is small: it makes mistakes, forgets earlier messages and works best in English, so check anything important. Powered by wllama (llama.cpp in WebAssembly); the browser builds are on Hugging Face.

The KaiLu family

Every KaiLu model is trained from random initialization. They borrow LFM2's hybrid layout of short-convolution and attention layers, not its weights.

  • Text GenerationNewestOpen source1.0B

    KaiLu 3 Pro

    KvcHub/KaiLu-3-Pro

    KaiLu 3 grown to 1B parameters and trained further on web, math and instruction data, then tuned with supervised fine-tuning and DPO. One 1.1 GB GGUF runs it in Ollama, llama.cpp and LM Studio. Start here.

    Parameters
    1.03B
    Layers
    36 (24 conv, 12 attn)
    Context
    1,024 tokens
    License
    Apache 2.0
  • Text Generation0.3B

    KaiLu 3

    KvcHub/KaiLu3

    The 329M base model that KaiLu 3 Pro grew from. Pretrained on 6.6B tokens of web, textbook, math and code data, then tuned with supervised fine-tuning and DPO. The smallest and fastest KaiLu.

    Parameters
    329M
    Layers
    24 (16 conv, 8 attn)
    Context
    1,024 tokens
    License
    Free for personal use
  • Text Generation2B · GGUF

    KaiLu 2 2B

    KvcHub/KaiLu-2-2B-GGUF

    The largest KaiLu, trained on a single Mac. It works through math, code and logic step by step before it answers.

    Parameters
    1.90B
    Layers
    104 (68 conv, 36 attn)
    Context
    1,024 trained, 4,096 max
    License
    Apache 2.0
  • Text Generation1.0B

    KaiLu 2 933M

    KvcHub/KaiLu-2-933M

    Built with MLX for Apple silicon, with a 546 MB GGUF build sized for phones.

    Parameters
    933M
    Speed
    260+ tok/s on Apple silicon
    Mobile RAM
    1.15 GB
    License
    Apache 2.0

All KaiLu models work best in English and can state wrong facts confidently, so they aren't meant for high-stakes use. KaiLu 3 Pro and KaiLu 2 are open source under Apache 2.0. KaiLu 3 (329M) is free for personal use, study and modification; for business use, email [email protected] first.

Run KaiLu on your machine

Pick a model and a runtime. Commands come straight from each model card.

Model
Runtime
Terminal
$ ollama run hf.co/KvcHub/KaiLu-3-Pro

Best in English, within 1,024 tokens of context, and at a low temperature such as 0.2, which the Ollama setup uses.

A quick look at Kava

Kava is the lab's pair of tiny token classifiers, at 636k and 2.01M parameters. They replace personal details with placeholders like [EMAIL] and [PHONE] before text is logged or sent to a language model. Try a browser version below; your text never leaves this page.

0 chars
Redacted output
Redacted text appears here.
Latency
–
Tokens
–
Memory (est.)
–
Sent over network
0 B
  • [EMAIL]0
  • [PHONE]0
  • [CARD]0
  • [IP]0

The browser demo uses lightweight pattern rules so it runs anywhere, with latency measured live on your device. The real Kava models do learned token classification in Python. Memory estimates assume fp16 weights plus the text buffers.

Install

pip install kava-privacy

Speed per request

KvcHub on-device inference<1 ms
Cloud DLP / LLM API round trip~130 ms+
Hatched time is spent on the network and in someone else's queue. Cloud figure is typical and varies by provider and region.