Instructions to use samdei/samdei-type-mini with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use samdei/samdei-type-mini with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "translation" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # pip install "transformers<5.0.0" from transformers import pipeline pipe = pipeline("translation", model="samdei/samdei-type-mini")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("samdei/samdei-type-mini") model = AutoModelForSeq2SeqLM.from_pretrained("samdei/samdei-type-mini", device_map="auto") - Notebooks
- Google Colab
- Kaggle
# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained("samdei/samdei-type-mini")
model = AutoModelForSeq2SeqLM.from_pretrained("samdei/samdei-type-mini", device_map="auto")Samdei Type Mini
Samdei Type Mini converts romanized Khmer, as people type it in chat, into Khmer script. It is a 7.1M-parameter character-level encoder-decoder, small enough for a phone keyboard: 7 MiB as int8 ONNX.
| Input | Output |
|---|---|
knhom jong tov phsar |
αααα»αα αααα ααααΆα |
ot mean wifi te |
α’ααααΆα wifi αα |
Release article: Samdei Type: open models that turn romanized Khmer into Khmer script, with a demo of this model in a chat.
Quick start
With transformers alone (pip install transformers torch):
from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
repo = "samdei/samdei-type-mini"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForSeq2SeqLM.from_pretrained(repo)
def pieces(text: str, max_words: int = 8) -> list[str]:
"""At most max_words words per piece, with a break after punctuation, as in the paper."""
out, cur = [], []
for w in text.split():
cur.append(w)
if len(cur) == max_words or w[-1] in ",.;:?!α":
out.append(" ".join(cur))
cur = []
if cur:
out.append(" ".join(cur))
return out or [text]
def convert(text: str) -> str:
batch = tok(pieces(text), padding=True, truncation=True, return_tensors="pt")
out = model.generate(**batch) # beam search, k = 5 (generation_config.json)
return " ".join(tok.batch_decode(out, skip_special_tokens=True))
print(convert("knhom jong tov phsar")) # αααα»αα
αααα
ααααΆα
This is beam search alone (CER 0.072 on Samdei Type Bench). For the best quality, rerank with the word LM of this repo (0.058), using the samdei-type package:
pip install "samdei-type[deploy]"
from samdei_type.pipeline import Converter
conv = Converter("samdei/samdei-type-mini") # int8 ONNX + word-LM rerank, as in the paper
print(conv.convert("knhom jong tov phsar")) # αααα»αα
αααα
ααααΆα
Or on the command line: samdei-type "knhom jong tov phsar".
Files
config.json, generation_config.json, model.safetensors (28.4 MiB), tokenizer*.json fp32 model (transformers)
onnx/encoder_kv.int8.onnx (5.0 MiB), onnx/decoder_step.int8.onnx (2.0 MiB), onnx/vocab.json int8 ONNX
lm/wordfull_nowiki_p25.json.gz (1.2 MiB) word bigram LM for reranking
rerank.json decoding and rerank settings
On a phone, the two ONNX graphs are all a runtime needs. encoder_kv runs once per input. decoder_step decodes one token with a KV cache, and its batch axis holds all beams. samdei_type/onnx_infer.py is a plain-numpy reference to port from. On one Apple M5 Pro CPU thread, beam 5 takes 4.9 ms per input (p50); with the rerank, a keystroke takes 5.7 ms. It has not been timed on a phone yet.
Results
Samdei Type Bench: 2,155 everyday messages and words with human-checked Khmer. The metrics are CER after normalization and exact match.
| System | CER | Exact |
|---|---|---|
| Samdei Type Mini, reranked | 0.058 | 24% |
| Samdei Type Mini, beam only (the transformers example above) | 0.072 | 17% |
| Samdei Type Base, reranked | 0.052 | 30% |
| Samdei Type Prime (not released) | 0.027 | 61% |
| Akkhara-mini, 6-word pieces (0.159 on whole inputs) | 0.066 | 23% |
| AkaraAlpha | 0.281 | 2% |
- Prime is Base trained further on scenario-designed, LLM-drafted, human-audited data. Most bench items share that scenario design, so its lead is an upper bound.
- Decode long messages in pieces, as above. On whole inputs, beam-only CER is 0.218.
Model and training
- Architecture. Pre-LN Transformer encoder-decoder: 6 encoder and 2 decoder layers, d_model 256, 4 heads, FFN 1024. It has 215 input symbols (Latin letters, digits, punctuation) and 228 output symbols (Khmer characters, and copied characters).
- transformers format. It loads as
BlenderbotForConditionalGenerationbecause the architecture is the same; it has nothing else in common with BlenderBot. Both vocabularies share one id space (output symbols first), so transformers counts 7.43M parameters. - Training. Three stages on one laptop GPU (Apple M5 Pro), about 3 hours in all:
- pretraining on 6.5M pairs: real Khmer sentences spelled by a rule-based typing generator (
samdei-type-mini-pretrained); - fine-tuning on 1.16M chat-like pairs;
- adaptation to casual register (1.59M pairs, with replay).
- pretraining on 6.5M pairs: real Khmer sentences spelled by a rule-based typing generator (
- Data. The exact training files are in
samdei-type-pretraining-dataandsamdei-type-finetuning-data, withrecipes.json. No evaluation data and no other romanization system's data was trained on. Akkhara-Preview, a public 110-row preview, was used only to pick stage 1β2 checkpoints; stage 3 used a closed development set.
Limitations
- Every romanized input seen in training is synthetic (rules, mined spelling tables, an LLM). Real typing appears only in the evaluation.
- Homophones that need more context than a word bigram: ααΎα / ααα (
der), ααΉα / ααα / ααα (dek). - Loanwords typed in English (
checkfor ααα) stay English. A Latin comma is copied, and a missing final α is never added. - Rare dictionary words without context are hard, as is heavier abbreviation than in training.
- The Khmer side is mostly web text, with its topics and biases.
- Not for Khmer β Latin, official romanization of names, or output nobody checks.
License
Apache-2.0 (LICENSE; keep NOTICE when redistributing).
- Word LM: built without Wikipedia, from HPLT v2, FineWeb-2 (ODC-By), ALT (CC BY) and our LLM-written chat sentences.
- Training data: CC BY-SA 4.0, with its sources in the dataset cards.
- Tools:
khmercut(Apache-2.0) and SILkhnormal(MIT).
Citation
@misc{samdei-type-2026,
title = {Samdei Type: Separating What Is Written from How It Is Spelled for Romanized Khmer Transliteration},
author = {Mengsay Loem},
year = {2026},
note = {Technical report},
url = {https://github.com/Samdei-Labs/samdei-type/blob/main/paper/samdei-type.pdf}
}
- Downloads last month
- 97
Model tree for samdei/samdei-type-mini
Base model
samdei/samdei-type-mini-pretrainedDatasets used to train samdei/samdei-type-mini
samdei/samdei-type-pretraining-data
Collection including samdei/samdei-type-mini
Evaluation results
- CER (beam 5, word-LM rerank, 8-word pieces) on Samdei Type Benchtest set self-reported0.058
- Exact match on Samdei Type Benchtest set self-reported0.239
# Use a pipeline as a high-level helper # Warning: Pipeline type "translation" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # pip install "transformers<5.0.0" from transformers import pipeline pipe = pipeline("translation", model="samdei/samdei-type-mini")