law-il-E2B-safetensors

bx-legal-kochav (safetensors)

Repository id: BrainboxAI/law-il-E2B-safetensors

The full 16-bit weights of the legal model law-il-E2B. This is a companion repository, not a separate product.

bx-legal-kochav is the model's name under the BrainboxAI naming convention. The repository id has not changed and will not change. Every existing link and script keeps working.

---

What is in here

One file: model.safetensors, 10.2 GB, alongside the tokenizer and the chat template.

These are the same weights as in the main repository, in a different format. The main repository holds a compressed build that runs on ordinary hardware. This is the full-precision build. It is larger, heavier, and meant for working on the model rather than just running it.

When to take this build

• Loading the model with transformers in Python.
• Continuing training on a firm's or a company's own material.
• Converting to another deployment format, such as ONNX.
• A framework that cannot read GGUF.

When not to take it

If all you want is to run the model and ask it questions, take the main repository, BrainboxAI/law-il-E2B. It is far smaller and works directly with Ollama or LM Studio. The file here will demand a great deal more memory and will not give you better answers.

How to load it

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("BrainboxAI/law-il-E2B-safetensors")
model = AutoModelForCausalLM.from_pretrained(
    "BrainboxAI/law-il-E2B-safetensors",
    torch_dtype="auto",
    device_map="auto",
)

messages = [
    # "I was fired without notice. What am I entitled to?"
    {"role": "user", "content": "פיטרו אותי בלי הודעה מוקדמת. מה מגיע לי?"},
]
inputs = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True)
outputs = model.generate(inputs, max_new_tokens=512, temperature=0.3, top_p=0.9)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

The prompt is in Hebrew because that is the language this model works in. The comment above it is the English gloss.

The full card lives elsewhere

Everything worth knowing before you rely on this model, what it was trained on, what it does not know, and what you must not do with it, is on the main repository's card:

BrainboxAI/law-il-E2B

Do not use the model without reading it. This is a legal model, and its limitations are the important part of it.

The repositories for this model

License

Apache 2.0.

This is a fine-tune of unsloth/gemma-4-E2B-it, so the terms of that model apply here as well. The base model is published under Apache 2.0 and also points to the Gemma 4 licence terms. Read those before commercial use.

Author

Built by Netanel Elyasi, founder of BrainboxAI, an Israeli applied-AI studio building small, private, domain-specialised models.

Questions, corrections, or a use case this model does not cover: [email protected].

Part of the BrainboxAI family of on-device models. See also code-il-E4B (code) and cyber-analyst-4B (security).