GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
aungkomyint
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
aungkomyint
Model Tree
aungkomyint/tara10m-sft-v1-2k
Input Modalities
Output Modalities
Supported Functionality
GPT2LMHeadModelsafetensorsThe reference Tara10M model is a 10.4M-parameter Llama-style Burmese-English SFT model with a 16K SentencePiece vocabulary and 1,024-token context. Tara 1.1 is not the same architecture and is not a drop-in replacement for Tara10M.
Tara 1.1 is the smaller educational branch:
Training was staged locally:
data/base_school.txt.data/sft_assistant.txt.data/multiturn_fix.txt.tara-1m-assistant-v2 to preserve stronger assistant behavior while adding domain knowledge.The final shipped checkpoint is:
text
outputs/tara-1m-assistant-v5-packedLower is better.
| Model | Eval file | Loss | Perplexity |
|---|---|---|---|
| tara-1m-assistant-v2 | data/sft_assistant.txt | 3.0114 | 20.3157 |
| Tara 1.1 1M SFT | data/sft_assistant.txt | 2.1528 | 8.6091 |
| tara-1m-assistant-v2 | data/multiturn_fix.txt | 4.0452 | 57.1240 |
| Tara 1.1 1M SFT |
Best test areas:
Prompt format:
text
User: What is SFT?Assistant:python
from transformers import AutoModelForCausalLM, AutoTokenizer repo_id = "aungkomyint/tara1.1" tokenizer = AutoTokenizer.from_pretrained(repo_id)model = AutoModelForCausalLM.from_pretrained(repo_id) prompt = "User: What is SFT?\nAssistant:"inputs = tokenizer(prompt, return_tensors="pt")outputs = model.generate( **inputs, max_new_tokens=80, do_sample=False, pad_token_id=tokenizer.pad_token_id or tokenizer.eos_token_id,) print(tokenizer.decode(outputs[0], skip_special_tokens=True))Local folder usage:
python
from transformers import pipeline pipe = pipeline("text-generation", model="./tara1.1-hf")print(pipe("User: What is Tara 1M?\nAssistant:", max_new_tokens=80)[0]["generated_text"])Prompt:
text
User: What is SFT?Assistant:Output:
text
SFT means supervised fine tuning. It trains the model on examples of instructions and good answers.Prompt:
text
User: How do I improve the model?Assistant:Output:
text
Use cleaner data, train longer, test outputs, then add better examples for weak areas.This model is intentionally tiny and should be treated as an experiment.
Known limitations:
model.safetensorsconfig.jsongeneration_config.jsontokenizer.jsontokenizer_config.jsonspecial_tokens_map.jsonvocab.jsonmerges.txtchat_template.jinjadata/multiturn_fix.txt |
| 3.2565 |
| 25.9596 |