Instructions to use rudrashah/RLM-hinglish-translator-bidir-2-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use rudrashah/RLM-hinglish-translator-bidir-2-gguf with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use rudrashah/RLM-hinglish-translator-bidir-2-gguf with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf rudrashah/RLM-hinglish-translator-bidir-2-gguf:F16 # Run inference directly in the terminal: llama cli -hf rudrashah/RLM-hinglish-translator-bidir-2-gguf:F16
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf rudrashah/RLM-hinglish-translator-bidir-2-gguf:F16 # Run inference directly in the terminal: llama cli -hf rudrashah/RLM-hinglish-translator-bidir-2-gguf:F16
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 rudrashah/RLM-hinglish-translator-bidir-2-gguf:F16 # Run inference directly in the terminal: ./llama-cli -hf rudrashah/RLM-hinglish-translator-bidir-2-gguf:F16
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 rudrashah/RLM-hinglish-translator-bidir-2-gguf:F16 # Run inference directly in the terminal: ./build/bin/llama-cli -hf rudrashah/RLM-hinglish-translator-bidir-2-gguf:F16
Use Docker
docker model run hf.co/rudrashah/RLM-hinglish-translator-bidir-2-gguf:F16
- LM Studio
- Jan
- Ollama
How to use rudrashah/RLM-hinglish-translator-bidir-2-gguf with Ollama:
ollama run hf.co/rudrashah/RLM-hinglish-translator-bidir-2-gguf:F16
- Unsloth Desktop
- Docker Model Runner
How to use rudrashah/RLM-hinglish-translator-bidir-2-gguf with Docker Model Runner:
docker model run hf.co/rudrashah/RLM-hinglish-translator-bidir-2-gguf:F16
- Lemonade
How to use rudrashah/RLM-hinglish-translator-bidir-2-gguf with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull rudrashah/RLM-hinglish-translator-bidir-2-gguf:F16
Run and chat with the model
lemonade run user.RLM-hinglish-translator-bidir-2-gguf-F16
List all available models
lemonade list
- Atomic Chat
RLM-Hinglish-to-English-Bidirectional β GGUF (Q8_0)
This is the Q8_0 quantized GGUF version of RLM-hinglish-translator-bidir-2, a fully bidirectional Hinglish β English translation model. Converted using Unsloth, this quantized version offers near full-precision quality with significantly reduced memory usage β making it easy to run locally with llama.cpp or Ollama.
What is This Model?
RLM-Hinglish-to-English-Bidirectional translates in both directions between Hinglish (the code-mixed blend of Hindi and English widely used across India) and English β all within a single model. No need for two separate models.
- β Hinglish β English
- β English β Hinglish
- β Chat-format, instruction-following
- β Optimized for short to medium informal sentences (everyday chat style)
Available File
| File | Quantization | Quality | Use Case |
|---|---|---|---|
RLM-hinglish-translator-bidir-2.Q8_0.gguf |
Q8_0 | Near lossless | Best quality, recommended |
Usage
llama.cpp
llama-cli -hf rudrashah/RLM-hinglish-translator-bidir-2-gguf --jinja
Ollama
Run directly:
ollama run rudrashah/RLM-hinglish-translator-bidir-2-gguf
Or create a local model from the GGUF file:
ollama create RLM-hinglish-bidir -f ./Modelfile
ollama run RLM-hinglish-bidir
Python (Transformers)
Simply prefix your text with Hinglish: or English: to set the translation direction.
from transformers import TextStreamer
# Hinglish β English
messages = [{
"role": "user",
"content": [{"type": "text", "text": "Hinglish: kal mujhe office jaana hai"}]
}]
# English β Hinglish
# messages = [{
# "role": "user",
# "content": [{"type": "text", "text": "English: I have to go to the office tomorrow."}]
# }]
inputs = tokenizer.apply_chat_template(
messages,
add_generation_prompt=True,
return_tensors="pt",
tokenize=True,
return_dict=True,
).to("cuda")
_ = model.generate(
**inputs,
max_new_tokens=128,
use_cache=True,
temperature=1.0, top_p=0.95, top_k=64,
streamer=TextStreamer(tokenizer, skip_prompt=True),
)
Note: Just change the prefix β use
Hinglish:to translate to English, orEnglish:to translate to Hinglish.
Examples
Hinglish β English
| Hinglish Input | English Output |
|---|---|
aapka name kya hai? |
What is your name? |
kal mujhe office jaana hai |
I have to go to the office tomorrow. |
Tumne kal ki match dekhi thi? kon man of the match bana tha? |
Did you watch yesterday's match? Who was man of the match? |
English β Hinglish
| English Input | Hinglish Output |
|---|---|
What is your name? |
aapka naam kya hai? |
I have to go to the office tomorrow. |
kal mujhe office jaana hai. |
Did you watch yesterday's match? |
Tumne kal ki match dekhi thi? |
Model Details
| Property | Details |
|---|---|
| Base Model | 4B instruction-tuned LLM |
| Fine-tuning Method | PEFT (LoRA), rank 64, lora_alpha 64 |
| Trainer | TRL SFTTrainer |
| Dataset | findnitai/english-to-hinglish |
| Training Samples | 15,000 (both directions) |
| Format | Chat template (user/assistant) |
| Quantization | Q8_0 via Unsloth |
| License | MIT |
Limitations
- Best suited for short to medium-length informal sentences, similar to everyday chat messages.
- May struggle with heavy regional slang, very long sentences, or dialect-specific expressions.
- English β Hinglish outputs reflect the Hinglish style present in the training data and may not match all regional variations.
Model Family
| Model | Direction | Format | Notes |
|---|---|---|---|
| RLM-hinglish-translator | Hinglish β English | Full precision | v1, 2B base |
| RLM-hinglish-translator-2 | Hinglish β English | Full precision | v2, 4B base |
| RLM-hinglish-translator-bidir-2 | Hinglish β English | Full precision | Bidirectional |
| RLM-hinglish-translator-bidir-2-gguf (this) | Hinglish β English | GGUF Q8_0 | Quantized for local use |
- Downloads last month
- 42
8-bit