Instructions to use kartikey31/txn-parser with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use kartikey31/txn-parser with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="kartikey31/txn-parser",
	filename="student/gguf/gemma-3-270m-it.BF16.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use kartikey31/txn-parser with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf kartikey31/txn-parser:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf kartikey31/txn-parser:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf kartikey31/txn-parser:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf kartikey31/txn-parser:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf kartikey31/txn-parser:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf kartikey31/txn-parser:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf kartikey31/txn-parser:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf kartikey31/txn-parser:Q4_K_M

Use Docker

docker model run hf.co/kartikey31/txn-parser:Q4_K_M

LM Studio
Jan

vLLM

How to use kartikey31/txn-parser with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "kartikey31/txn-parser"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "kartikey31/txn-parser",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/kartikey31/txn-parser:Q4_K_M

Ollama
How to use kartikey31/txn-parser with Ollama:
```
ollama run hf.co/kartikey31/txn-parser:Q4_K_M
```

Unsloth Studio new

How to use kartikey31/txn-parser with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for kartikey31/txn-parser to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for kartikey31/txn-parser to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for kartikey31/txn-parser to start chatting

Docker Model Runner
How to use kartikey31/txn-parser with Docker Model Runner:
```
docker model run hf.co/kartikey31/txn-parser:Q4_K_M
```

Lemonade

How to use kartikey31/txn-parser with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull kartikey31/txn-parser:Q4_K_M

Run and chat with the model

lemonade run user.txn-parser-Q4_K_M

List all available models

lemonade list

Transaction Parser — Voice → JSON (on-device)

Distilled student model that turns voice-transcribed transaction strings into structured JSON for an Android expense-tracking app. Examples:

Input	Output
`"500 rs on beer 50 rs on candy"`	`[{amount: 500, item: "beer", category: "Drinks", ...}, {amount: 50, item: "candy", ...}]`
`"do sau rupay ka chai"`	`[{amount: 200, currency: "INR", item: "chai", category: "Drinks", ...}]`
`"1.5k for shoes from myntra"`	`[{amount: 1500, item: "shoes", category: "Shopping", ...}]`
`"got my salary 50000"`	`[{amount: 50000, type: "income", category: "Income", ...}]`

What's in this repo

Path	Description
`student/gguf/gemma3_text-fixed.BF16.gguf`	Lossless ref (543 MB)
`student/gguf/gemma3_text-fixed.Q8_0.gguf`	High quality (~290 MB)
`student/gguf/gemma3_text-fixed.Q5_K_M.gguf`	Default for ship (260 MB)
`student/gguf/gemma3_text-fixed.Q4_K_M.gguf`	Smallest but lossy on 270M (253 MB)
`student/adapters/`	Trained LoRA adapter (r=32, α=64) for further finetuning
`teacher/gguf/gemma-4-e2b-it.Q3_K_M.gguf`	Teacher (Gemma 4 E2B) used for distillation labeling
`teacher/adapters/`	Teacher LoRA adapter (r=16, α=32)

Recommended file

student/gguf/gemma3_text-fixed.Q5_K_M.gguf — 260 MB, 94% JSON valid, runs on-device on Android via llama.cpp at ~150 ms per request on a modern mid-range device.

Evaluation (50-example smoke test)

Build	Size	JSON valid	Schema valid	Exact match (numeric-aware)	Mean latency (A100)
fp16 adapter (ceiling)	n/a	98%	94%	~48%	1219 ms
BF16 GGUF (fixed)	543 MB	98%	74%	48%	108 ms
Q8_0 GGUF (fixed)	~290 MB	~98%	~74%	~46%	~120 ms
Q5_K_M GGUF (fixed)	260 MB	94%	72%	20%	210 ms
Q4_K_M GGUF (fixed)	253 MB	68%	56%	18%	177 ms

The "exact-match" column uses numeric-aware comparison (100 == 100.0). Most "schema invalid" failures are missing-field or enum-value drift; the category prediction is mostly diagonal in the confusion matrix.

Tip for Android: always run a JSON.parse → schema validate → fallback UI pipeline. ~6% of inputs at Q5_K_M will fail to parse — handle that as "couldn't understand, please try again" rather than crashing.

Usage

`llama.cpp` / `llama-cpp-python` (Python)

from llama_cpp import Llama

llm = Llama(
    model_path="gemma3_text-fixed.Q5_K_M.gguf",
    n_gpu_layers=-1,
    n_ctx=2048,
    verbose=False,
)

SYSTEM_PROMPT = (
    "You convert short, possibly code-mixed (English/Hindi/Hinglish) "
    "transcribed transaction strings into a JSON object with a single "
    '"transactions" array. Each transaction has: amount (number), '
    "currency (string, default 'INR'), item (string), category (one of "
    "Food, Drinks, Groceries, Transport, Shopping, Entertainment, Bills, "
    "Health, Education, Personal, Gifts, Income, Other), type "
    "('expense' or 'income'). Output ONLY the JSON object — no prose."
)

resp = llm.create_chat_completion(
    messages=[
        {"role": "system", "content": SYSTEM_PROMPT},
        {"role": "user", "content": "500 rs on beer 50 rs on candy"},
    ],
    temperature=0.0, top_p=1.0, max_tokens=512,
)
print(resp["choices"][0]["message"]["content"])

Android (`llama.cpp` JNI)

Bundle gemma3_text-fixed.Q5_K_M.gguf in your app (or download on first run).
Use the llama.cpp Android example or a JNI wrapper.
Set the same system prompt above; user message = the voice transcript.
Validate output with a JSON-schema library on the parse path.

Keep the llama_context alive across requests — don't reload per call.

Quick test on Linux/macOS

huggingface-cli download kartikey31/txn-parser \
    --repo-type=model --local-dir models

python -c "
from llama_cpp import Llama
llm = Llama(model_path='models/student/gguf/gemma3_text-fixed.Q5_K_M.gguf', n_gpu_layers=-1, verbose=False)
print(llm.create_chat_completion(messages=[
    {'role':'system','content':'Output only JSON with a transactions array...'},
    {'role':'user','content':'500 rs on beer 50 rs on candy'},
], temperature=0)['choices'][0]['message']['content'])
"

Training details

Base model: unsloth/gemma-3-270m-it
Method: QLoRA via Unsloth (r=32, α=64, dropout 0.0, all linear targets)
Train data: 29,890 teacher-labeled examples (data/distill/train.jsonl) generated by a fine-tuned Gemma 4 E2B teacher
Epochs: 2
Effective batch: 128 (A100) / 16 (5060 Ti)
Optimizer: AdamW 8-bit, cosine LR, peak 2e-4, warmup 3%
Final eval loss: 0.099 (eval set: 300 hand-curated examples)
GGUF conversion: raw llama.cpp/convert_hf_to_gguf.py (NOT Unsloth's wrapper), preserves BOS token in chat template
Hardware: A100-SXM4-80GB, ~25 min total training time at batch 128

Code, dataset generation, evaluation, and conversion scripts: https://github.com/kartikeychoudhary/txn-parser

Categories enum

Food, Drinks, Groceries, Transport, Shopping, Entertainment, Bills, Health, Education, Personal, Gifts, Income, Other

License

Apache-2.0 (matches base model). The training data is synthetic and released under the same license.

Citation

@software{txn-parser-2026,
  author  = {Kartikey Choudhary},
  title   = {Transaction Parser: Voice-to-JSON distilled model},
  year    = {2026},
  url     = {https://huggingface.co/kartikey31/txn-parser},
  note    = {Gemma 3 270M, distilled from Gemma 4 E2B teacher},
}

Downloads last month: 270

GGUF

Model size

0.3B params

Architecture

gemma3

Hardware compatibility

3-bit

4-bit

5-bit

8-bit

16-bit

Model tree for kartikey31/txn-parser

Base model

google/gemma-3-270m

Finetuned

google/gemma-3-270m-it

Finetuned

unsloth/gemma-3-270m-it

Quantized

(29)

this model

Evaluation results

JSON valid (50-example eval)
self-reported

94.000
Schema valid
self-reported

72.000
Exact match (numeric-aware)
self-reported

20.000