| --- |
| license: apache-2.0 |
| language: |
| - en |
| - hi |
| - gu |
| - ta |
| - bn |
| - mr |
| tags: |
| - sumeru |
| - education |
| - ncert |
| - jee |
| - neet |
| - india |
| pipeline_tag: text-generation |
| --- |
| |
| # Sumeru-rm |
|
|
| **Sumeru-rm** (Research Mini) is a compact science and mathematics model from |
| **Sumeru AI**, purpose-built for the Indian education curriculum: NCERT |
| Classes 6-12 plus JEE and NEET preparation. Subjects: Physics, Chemistry, |
| Biology, Mathematics. |
|
|
| ## Usage |
|
|
| ```python |
| from transformers import AutoModelForImageTextToText, AutoProcessor |
| |
| processor = AutoProcessor.from_pretrained("debowd/sumeru-rm", trust_remote_code=True) |
| model = AutoModelForImageTextToText.from_pretrained( |
| "debowd/sumeru-rm", trust_remote_code=True, device_map="auto" |
| ) |
| |
| messages = [{"role": "user", "content": [{"type": "text", |
| "text": "[Physics | Class 11 | Laws of Motion | Easy]\n\nState Newton's second law."}]}] |
| text = processor.apply_chat_template(messages, add_generation_prompt=True, tokenize=False) |
| inputs = processor(None, text, return_tensors="pt").to(model.device) |
| output = model.generate(**inputs, max_new_tokens=300) |
| print(processor.tokenizer.decode(output[0], skip_special_tokens=True)) |
| ``` |
|
|
| ## Architecture |
|
|
| - Class: `SumeruForCausalLM` |
| - Model type: `sumeru` |
| - Loaded via `trust_remote_code=True` |
|
|
| ## License |
|
|
| Apache 2.0. Copyright (c) 2026 Sumeru AI. |
|
|
| ## Developed by |
|
|
| [Sumeru AI](https://huggingface.co/debowd) — Indian education AI. |
|
|
