Sansar 700M

README.md

@@ -17,17 +17,20 @@ We build language technology for **Sanskrit** and other Indic languages: corpora
 
 | Model | Parameters | Bits per byte (held-out, ex-Gītā) ↓ |
 |---|---|---|
-| [sansar-350m](https://huggingface.co/MuseMesh/sansar-350m) | 318M | **0.5547** |
+| [sansar-700m](https://huggingface.co/MuseMesh/sansar-700m) | 704M | **0.5307** |
+| [sansar-350m](https://huggingface.co/MuseMesh/sansar-350m) | 318M | 0.5547 |
 | [sansar-125m](https://huggingface.co/MuseMesh/sansar-125m) | 97M | 0.6039 |
 | [sansar-60m](https://huggingface.co/MuseMesh/sansar-60m) | 63M | 0.6434 |
 | [sansar-20m](https://huggingface.co/MuseMesh/sansar-20m) | 27M | 0.7177 |
 
 Scores are measured on frozen held-out Sanskrit sets: DCS gold sentences, prose, Vedic and out-of-domain texts. Held-out text is masked out of training. Every model card gives the per-set numbers, the training data and a version history.
 
+Scored the same way, general base models need more bits per byte: Gemma 3 4B 0.6965, Qwen3-4B 0.7071, Llama 3.2 3B 0.7122, Sarvam-1 0.7465. sansar-700m beats all four on every held-out set with 704M parameters (Krutrim-2 12B, scored earlier in a pass that is not strictly comparable, is not in this list). A live demo is at **[mume.ai/sansar](https://mume.ai/sansar)**.
+
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
-tok = AutoTokenizer.from_pretrained("MuseMesh/sansar-350m", trust_remote_code=True)
-model = AutoModelForCausalLM.from_pretrained("MuseMesh/sansar-350m", trust_remote_code=True)
+tok = AutoTokenizer.from_pretrained("MuseMesh/sansar-700m", trust_remote_code=True)
+model = AutoModelForCausalLM.from_pretrained("MuseMesh/sansar-700m", trust_remote_code=True)
 ids = tok("संस्कृतं नाम दैवी वाक्", return_tensors="pt")
 print(tok.decode(model.generate(**ids, max_new_tokens=60, do_sample=True, top_k=100)[0]))
 ```