# Udgam Hinglish Commands 0.6B

Canonical: https://udgamlabs.com/models/hinglish-commands · Updated: 2026-11-14

Models · Research release · 22 September 2026

Turns short English and Romanized Hinglish requests — Hindi and English mixed in Latin letters — into a structured intent and labelled details that an app can use.

[Read the release post](https://udgamlabs.com/blog/hinglish-commands)

The result

Complete-command exact match rose from 9.90% to 54.88%. 5,877 of 13,026 are still wrong.

Measured on the frozen Hinglish-TOP test against the original Qwen3-0.6B, both given the same prompt with four retrieved training examples. A command counts only if the whole structure matches the human reference: every label, value and nesting. The gain is 44.98 percentage points (paired 95% interval 43.93–46.03). It is one parsing task, not a measure of general Hindi or English ability.

## What it is

- **Type**: A LoRA adapter (fine-tune) on Qwen3-0.6B; not a newly pretrained model
- **Trained**: About 10.1 million adapter parameters; the Qwen3-0.6B weights stay unchanged
- **Input**: A short request in English or Romanized Hinglish (Hindi and English in Latin letters)
- **Output**: An intent with labelled details, as a TOP tree; the included code turns it into nested JSON
- **Domains**: Alarms, Events, Messaging, Music, Navigation, Reminders, Timers, Weather
- **Training data**: Human-labelled pairs from Hinglish-TOP, which builds on TOPv2; no synthetic data
- **Licence**: Adapter CC BY-SA 4.0; Udgam-written code Apache 2.0; Qwen materials keep Apache 2.0

A recorded example

bed time ka alarm kab hai? becomes the intent “Look up an alarm” with the detail alarm name: bed time. It proposes an interpretation and executes nothing; your app resolves dates and names and decides whether to act.

## Measured results

Hinglish-TOP test · complete-tree exact match

| Commands | Rows | Qwen3-0.6B | Adapter |
| --- | --- | --- | --- |
| All | 13,026 | 9.90% | 54.88% |
| Romanized Hinglish | 6,513 | 7.54% | 46.84% |
| English | 6,513 | 12.27% | 62.92% |
| Alarms | 2,390 | 16.15% | 61.63% |
| Events | 922 | 7.81% | 55.97% |
| Messaging | 1,012 | 9.98% | 63.83% |
| Music | 1,446 | 8.64% | 53.46% |
| Navigation | 1,996 | 9.22% | 42.74% |
| Reminders | 1,930 | 2.64% | 36.37% |
| Timers | 1,386 | 9.38% | 54.55% |
| Weather | 1,944 | 12.40% | 73.56% |

The test has 6,513 entries, each in English and in Romanized Hinglish, so the rows are paired, not independent. All eight domains improved, but the spread is wide: reminders are weakest at 36.37% and weather strongest at 73.56%.

[Read the full benchmark results](https://github.com/udgamlabs/hinglish-commands-0.6b/blob/main/docs/BENCHMARKS.md)

## Run it

Install the Python wrapper from the GitHub repository, download the adapter package from Hugging Face, and call the parser on a request. It needs the pinned Qwen3-0.6B base. The package includes the training examples it retrieves from; loading the weights alone with another prompt does not reproduce the measured setup.

Intended for developers and researchers testing a small local command parser. Have the user confirm a proposed command before your app acts on it.

## Scope and limitations

- 5,877 of 13,026 complete interpretations are still wrong, and Hinglish stays below half at 46.84%.
- Valid is not correct: 99.96% of outputs are well-formed trees, but a well-formed tree can still mean the wrong thing.
- Not native-script (Devanagari) Hindi or other languages, not speech, not general chat, and not shown to be reliable in production or fast on low-end phones.
- The results hold for this adapter with its four-retrieved-example prompt. Requests outside the eight domains can be forced into one of them.
- Public benchmark examples may have been in Qwen3-0.6B’s pretraining data, and the dataset, with its international names and places, is not a representative sample of everyday requests.

The adapter and prepared data are CC BY-SA 4.0, following the Hinglish-TOP and TOPv2 data it comes from. The original model and dataset authors have not endorsed this adaptation.

## Adapter, code and documentation

- [Hugging Face](https://huggingface.co/udgamlabs/hinglish-commands-0.6b)

    The adapter package and model card

- [GitHub repository](https://github.com/udgamlabs/hinglish-commands-0.6b)

    The Python wrapper, docs and licences

- [Full benchmark results](https://github.com/udgamlabs/hinglish-commands-0.6b/blob/main/docs/BENCHMARKS.md)

    Every table, interval and recorded failure

- [Visual walkthrough](https://udgamlabs.github.io/hinglish-commands-0.6b/)

    Recorded examples in plain language and JSON

---

## About Udgam Labs

Udgam Labs is an independent AI research lab. We work from first principles: before we build, we understand why things work, whether we are training a model, fine-tuning one or applying AI to a new problem. Today that means open models such as Gargi, a 384M-parameter language model that comes within 2 points of Qwen2.5-0.5B on three reasoning benchmarks with about 1,000× less training data. Next come products, custom models for businesses, and AI in the physical world. Udgam (उद्गम) is Sanskrit for origin.

Udgam Labs · AI research from first principles · https://udgamlabs.com