Instructions to use finndot/finnai-slm-v4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use finndot/finnai-slm-v4 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="finndot/finnai-slm-v4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("finndot/finnai-slm-v4")
model = AutoModelForCausalLM.from_pretrained("finndot/finnai-slm-v4", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

LiteRT

How to use finndot/finnai-slm-v4 with LiteRT:

# No code snippets available yet for this library.

# To use this model, check the repository files and the library's documentation.

# Want to help? PRs adding snippets are welcome at:
# https://github.com/huggingface/huggingface.js

LiteRT-LM

How to use finndot/finnai-slm-v4 with LiteRT-LM:

# LiteRT-LM runs on various platforms (Android, iOS, Windows, Linux, macOS, IoT, Web/WASM)
# and supports many APIs (C++, Python, Kotlin, Swift, JavaScript, Flutter).
# For platform-specific integration guides, please refer to the official developer website:
# https://ai.google.dev/edge/litert-lm

# To try LiteRT-LM, the easiest way is to use our CLI tool.
# 1. Install the LiteRT-LM CLI tool:
pip install -U litert-lm

# 2. Download and run this model locally:
# See: https://ai.google.dev/edge/litert-lm/cli
# A single .litertlm file in the repo is picked automatically; otherwise the CLI asks which one to run
# (or pass its name right after the repo id).
litert-lm run \
  --from-huggingface-repo=finndot/finnai-slm-v4 \
  --prompt="Write me a poem"

PEFT
How to use finndot/finnai-slm-v4 with PEFT:
```
Task type is invalid.
```
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use finndot/finnai-slm-v4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "finndot/finnai-slm-v4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "finndot/finnai-slm-v4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/finndot/finnai-slm-v4

SGLang

How to use finndot/finnai-slm-v4 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "finndot/finnai-slm-v4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "finndot/finnai-slm-v4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "finndot/finnai-slm-v4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "finndot/finnai-slm-v4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use finndot/finnai-slm-v4 with Docker Model Runner:
```
docker model run hf.co/finndot/finnai-slm-v4
```

FinnAI SLM v4

A 1.7B on-device language model for Indian bank SMS → JSON, Ask/Learn finance coaching, and Analytics tips.

Field	Value
Model ID	`finndot/finnai-slm-v4`
Base	`Qwen/Qwen3-1.7B`
Method	Continual QLoRA SFT from FinnAI v3 merged
Dataset	`finndot/finnai-slm-data-v4`
On-device INT4	`finndot/finnai-slm-v4-litertlm` (~929 MB `.litertlm`)
License	Apache 2.0
Eval	SHIP YES — see `eval/v20261005-finnai-slm-v4.md`

Built for FinnDot — private, on-device expense tracking. No user SMS leaves the phone.

Discovery / when to use this model

Use FinnAI SLM v4 if you are building:

Indian bank SMS → structured JSON (UPI, NEFT, IMPS, CC, salary, ATM)
On-device / mobile finance assistants (LiteRT-LM INT4 path)
Ledger-grounded Ask, Learn (tutor), or Analytics tips chat
Continual fine-tunes from a strong SMS+finance base (adapter/ included)

What it does

Mode	Description
SMS → JSON	`{amount, merchant, type, account, balance, category}` — empty `{}` for OTP/promo/KYC
Ask	Short answers grounded in a synthetic/user ledger
Learn	Tutor-style explanations (SIP, EMI, tax, budgeting)
Analytics	Spending-pattern tips without inventing numbers

Languages: English, Hindi, Hinglish, Tamil, Telugu, Marathi, Bengali (+ more Indic mix in train data).

Key results (frozen test)

Metric	FinnAI v4	Qwen3-1.7B base	Qwen2.5-1.5B
SMS R-EM % (n=2028)	94.87	29.39	49.9
Amount EM %	99.7	78.75	81.95
JSON valid %	100.0	100.0	99.65
Chat groundedness % (n=78)	80.77	47.44	65.38
False-parse %	0.25	100.0	12.88

Protocol: docs/llm-eval-protocol.md. Val SMS R-EM during train: 98.24%.

Quick start (Transformers)

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "finndot/finnai-slm-v4"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id, torch_dtype="bfloat16", device_map="auto"
)

messages = [
    {
        "role": "system",
        "content": (
            "Extract transaction details from this SMS as JSON: "
            "{amount, merchant, type, account, balance, category}. "
            "type is one of: INCOME, EXPENSE, CREDIT, TRANSFER, INVESTMENT. "
            "If this is not a transaction (OTP, promo, KYC), return {}."
        ),
    },
    {
        "role": "user",
        "content": "HDFC Bank: Rs.499.00 debited from A/c XX1234 to SWIGGY via UPI. Avl Bal Rs.15000.50",
    },
]
text = tokenizer.apply_chat_template(
    messages, tokenize=False, add_generation_prompt=True, enable_thinking=False
)
inputs = tokenizer(text, return_tensors="pt").to(model.device)
out = model.generate(**inputs, max_new_tokens=256, do_sample=False)
print(tokenizer.decode(out[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))

Load LoRA-only: adapter/ in this repo (PEFT), base Qwen/Qwen3-1.7B or prior FinnAI merged.

Training details

Parameter	Value
Base for continual train	FinnAI v3 merged (from Qwen3-1.7B)
Method	QLoRA SFT (4-bit NF4, bf16 compute)
LoRA r / α / dropout	16 / 32 / 0.05
Targets	q, k, v, o, gate, up, down_proj
Epochs / LR	3 / 1.5e-4 cosine, 3% warmup
Effective batch	16 (2 × 8)
Max seq	1536
Train size	40,000 (SMS 20k, chat 16.8k, general 3.2k)
Thinking	Disabled (`enable_thinking=false`)
Hardware	1× NVIDIA A10G (g5.2xlarge, us-east-2)
Seed	42

Privacy: fully synthetic — no user SMS, no real accounts.

On-device (LiteRT-LM)

Production file: finndot/finnai-slm-v4-litertlm

Format: .litertlm, INT4 dynamic_int4_block32, KV cache 1280, nothink
Size: 973,979,088 bytes
SHA256: 36e5ec021730f0e060b78c2371d1e0011c34f28c407cdf9f993232b5cd87e7c0

Limitations

Optimized for Indian banking SMS formats; weaker on US/EU-only banks (some hard cases remain: FAB, Schwab, etc.).
Merchant field remains hardest; regex parser-core still preferred when a bank parser exists.
Chat can still paraphrase; groundedness is improved but not perfect.
1.7B — task-specialized, not a general assistant.
On-device latency gate is measured separately on device hardware.

Reproduce

git clone https://github.com/AnomaProAI/Finndot
cd Finndot/ml
# dataset build + train configs: train/train_config_v4.yaml, data/out_v4_diverse/
python train/sft_qlora.py --config train/train_config_v4.yaml
python -m eval.run_eval --backend hf \
  --sms data/out_v4_diverse/test_sms.jsonl \
  --chat data/out_v4_diverse/chat_eval.jsonl \
  --model-id train/output/v4/merged \
  --baseline-id Qwen/Qwen3-1.7B \
  --tag v20261005-finnai-slm-v4

Citation

@misc{finnai-slm-v4-2026,
  title={FinnAI SLM v4: App-grounded on-device Indian bank SMS parsing and finance coaching},
  author={FinnDot Team},
  year={2026},
  url={https://huggingface.co/finndot/finnai-slm-v4},
  note={Apache 2.0. Continual QLoRA on Qwen3-1.7B / FinnAI v3.}
}

License

Apache 2.0 — same as Qwen3-1.7B.

Downloads last month: 230

Safetensors

Model size

2B params

Tensor type

BF16

Model tree for finndot/finnai-slm-v4

Base model

Qwen/Qwen3-1.7B-Base

Finetuned

Qwen/Qwen3-1.7B

Finetuned

(1275)

this model

Adapters

1 model

Finetunes

1 model

Dataset used to train finndot/finnai-slm-v4

Evaluation results

Strict Record Exact-Match (R-EM) on FinnAI v4 SMS test (synthetic, 2028)
self-reported

94.870
Amount Exact-Match on FinnAI v4 SMS test (synthetic, 2028)
self-reported

99.700
JSON Validity on FinnAI v4 SMS test (synthetic, 2028)
self-reported

100.000
Merchant Exact-Match on FinnAI v4 SMS test (synthetic, 2028)
self-reported

98.960
Groundedness on FinnAI v4 chat_eval (78)
self-reported

80.770