slm-500m-sft

slm-500m-sft

500M · QA SFT — answers a question from a supplied passage.

One of 13 models in a controlled study of how far a small language model can be pushed on US legal and financial text. Every version was trained on the same data, evaluated on the same frozen held-out set, and scored by the same blind LLM judge, so the stages are directly comparable. Compare them side by side in the SLM Arena.

Trained and aligned by Harman Sandhu (Vizuara AI Labs).

At a glance

Usage

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

MODEL = "Ace-2504/slm-500m-sft"
model = AutoModelForCausalLM.from_pretrained(MODEL, torch_dtype=torch.bfloat16).eval()
tok = AutoTokenizer.from_pretrained(MODEL)

SYS = ('You are a precise legal and financial assistant. Answer clearly using the '
       'provided context; do not invent facts.')
user = 'Context:\n<passage>\n\nQuestion: <your question>'

# Custom chat scheme. NOTE: never prepend <|bos|> — it was left untrained.
prompt = f'<|system|>\n{SYS}<|eos|>\n<|user|>\n{user}<|eos|>\n<|assistant|>\n'
enc = tok(prompt, return_tensors='pt', add_special_tokens=False).to(model.device)
eos = tok.convert_tokens_to_ids('<|eos|>')
out = model.generate(**enc, max_new_tokens=160, eos_token_id=eos,
                     pad_token_id=eos)   # pad and eos share an id here
print(tok.decode(out[0, enc['input_ids'].shape[1]:], skip_special_tokens=True))

Evaluation

This checkpoint was not evaluated. The published study evaluated the base, DPO and RLAIF legs of the 500M line; this intermediate checkpoint was trained and released for completeness but never scored, so no numbers are quoted here rather than borrowed from a sibling model.

Architecture

Training

• Initialised from thesreedath/slm-500m-base
• Method — full fine-tune
• Data — 15,000 grounded QA pairs, generated from a legal/financial corpus and gated by an LLM judge for faithfulness
• Schedule — 3 epochs · 2,812, 31.6 min on L4
• Loss is masked to the answer tokens only, so the model is never trained to reproduce the prompt

Cost — $1.03. The base was imported, so only this project's own work is counted: $1.03 of Modal GPU time and shared dataset generation.

Limitations and intended use

• Not legal or financial advice. This is a research artefact for studying small-model training, not a professional tool. Do not rely on its output for real decisions.
• At 517.8M it holds very little world knowledge. It is built to read an answer out of a passage you supply, not to recall facts. Used closed-book it will produce fluent, confident and wrong text.
• The judge behind these scores has not been calibrated against human labels, so treat small differences between models cautiously.
• English only; the corpus is US case law, SEC filings and educational web text.

The rest of the family

All under Ace-2504 except the two imported bases.

Citation

@misc{sandhu2026slm,
  title  = {Small Language Models for Legal and Financial Text: a controlled study of
            pretraining, instruction tuning, retrieval augmentation and alignment},
  author = {Harman Sandhu},
  year   = {2026},
  note   = {https://slm-arena-harman.vercel.app}
}