---
language:
- zh
- vi
license: cc-by-4.0
library_name: transformers
pipeline_tag: translation
tags:
- translation
- chinese-vietnamese
- marian
- webnovel
- xianxia
- nmt
- machine-translation
- asymmetric-transformer
- trained-from-scratch
datasets:
- ngocdang83/tran-vi-teacher
- chi-vi/hirashiba-mt-zh2vi-b-filtered
---

# HachimiMT-60: Chinese→Vietnamese Web-Novel Translation Model

A **56.94M-parameter Marian-class** Chinese-to-Vietnamese translation model
optimized for **web-novel content** (xianxia, modern, cross-domain).

## TL;DR

| Aspect | Value |
|---|---|
| **Params** | 56.94M |
| **Architecture** | Asymmetric Marian (8 encoder + 2 decoder, d_model 512) |
| **Vocab** | Custom SPM-BPE 24k joint ZH+VI |
| **Max position** | 512 |
| **Best for** | Xianxia + cross-domain web-novel paragraph translation |

## Quick Start

```python
from transformers import AutoTokenizer, MarianMTModel
import torch

tokenizer = AutoTokenizer.from_pretrained("ngocdang83/HachimiMT-60-zh-vi")
model = MarianMTModel.from_pretrained("ngocdang83/HachimiMT-60-zh-vi").to("cuda").eval()

src = "他必须得抓紧时间了。凌伊山掏出手机，查询起了临江市最近开往雪霏市的机票。"
inp = tokenizer(src, return_tensors="pt", truncation=True, max_length=256).to("cuda")
with torch.inference_mode():
    out = model.generate(
        **inp,
        max_new_tokens=300,
        num_beams=4,
        early_stopping=True,
        no_repeat_ngram_size=2,
        repetition_penalty=1.2,
    )
print(tokenizer.decode(out[0], skip_special_tokens=True))
# Output: "Hắn phải tranh thủ thời gian rồi. Lăng Y Sơn lấy điện thoại ra, tra
#  vé máy bay gần nhất từ thành phố Lâm Giang đến thành phố Tuyết Phi."
```

## Fast CPU Runtime

This repository also includes a CTranslate2 INT8 export under
`ct2-int8_float32/`, used by the public demo Space for faster CPU inference.

```python
import ctranslate2
from pathlib import Path
from huggingface_hub import snapshot_download
from transformers import AutoTokenizer

model_id = "ngocdang83/HachimiMT-60-zh-vi"
model_path = Path(snapshot_download(model_id, allow_patterns=[
    "config.json", "source.spm", "target.spm", "vocab.json", "tokenizer_config.json",
    "ct2-int8_float32/*",
]))

tokenizer = AutoTokenizer.from_pretrained(model_path)
translator = ctranslate2.Translator(
    str(model_path / "ct2-int8_float32"),
    device="cpu",
    compute_type="int8_float32",
)
```

## Speed Benchmark

Tested on RTX 5070 Ti Laptop, `num_beams=4`, mixed test set (20 short + 20 medium + 20 long rows).

| Model | Params | Mean Latency | max_position | Notes |
|---|---:|---:|---:|---|
| [Hirashiba-tiny](https://huggingface.co/chi-vi/hirashiba-mt-tiny-zh-vi) | 15.1M | **377ms** | 512 | Fastest |
| [Hirashiba-medium](https://huggingface.co/Moleys/hirashiba-mt-medium) | 57.07M | 495ms | 128 | Truncates paragraphs |
| **HachimiMT-60** (this) | 56.94M | **603ms** | 512 | Handles long paragraph without truncation |

Per-bucket mean latency (ms):

| Bucket | HachimiMT-60 | Hirashiba-medium | Hirashiba-tiny |
|---|---:|---:|---:|
| short (~70-120ch) | 330 | 390 | 310 |
| medium (~150-250ch) | 626 | 546 | 430 |
| long (>250ch) | 853 | 548 | 390 |

⚠️ **Hirashiba-medium and Hirashiba-tiny truncate** on medium/long buckets due
to `max_position_embeddings=128`, which caps output to ~120 tokens regardless
of source length. Their lower latency on long bucket reflects truncated output
rather than faster decoding. HachimiMT-60 produces full-length output up to
~1000 chars without truncation.

For ultra-low-latency short-content use cases, consider Hirashiba-tiny.
For paragraph-level web-novel translation, HachimiMT-60 is recommended.

## Architecture

```
MarianMTModel:
  vocab_size: 24000
  d_model: 512
  encoder_layers: 8
  decoder_layers: 2
  encoder_attention_heads: 8
  decoder_attention_heads: 8
  encoder_ffn_dim: 3072
  decoder_ffn_dim: 3072
  max_position_embeddings: 512
  share_encoder_decoder_embeddings: true
  tie_word_embeddings: true
  scale_embedding: true
  activation_function: swish
```

Total params: **56,935,424 (~57M)**.

## Training Datasets

Primary training sources:

1. **[ngocdang83/tran-vi-teacher](https://huggingface.co/datasets/ngocdang83/tran-vi-teacher)** —
   350k strict-clean Chinese-Vietnamese parallel from Gemini 2.5/3.0/3.1
   teacher (Pro/Flash/Flash-Lite tiers). Provides paragraph-level training
   examples + cross-domain coverage (urban, fantasy, sci-fi, history).

2. **[chi-vi/hirashiba-mt-zh2vi-b-filtered](https://huggingface.co/datasets/chi-vi/hirashiba-mt-zh2vi-b-filtered)** —
   Filtered Chinese-Vietnamese translation dataset for web-novel domain.

3. **Gold teacher** generated by Gemini API for additional quality-targeted
   training examples.

## Decode Configuration

Recommended generation parameters:

```python
out = model.generate(
    **inputs,
    max_new_tokens=300,       # adjust based on expected length
    num_beams=4,              # quality/speed tradeoff
    early_stopping=True,
    no_repeat_ngram_size=2,   # prevent repetition
    repetition_penalty=1.2,
)
```

For shorter inputs (single sentence), reduce `max_new_tokens=150`.
For long paragraphs, increase to `400`.

## Intended Uses

### Recommended

1. **Chinese-Vietnamese web-novel translation** (xianxia, tu tiên, fantasy, sci-fi)
2. **Paragraph-level translation** (handles up to ~1000 chars output without truncation)
3. **Cross-domain content** (Lovecraftian, urban, school, military)
4. **Production deployment** with batched inference

### Not Recommended

1. **Non-Chinese sources** (ZH→VI only, not bidirectional)
2. **Traditional Chinese (繁體) input** — model trained on Simplified
   Chinese (简体). Traditional characters may degrade output quality;
   convert to Simplified first (e.g. via `opencc`).
3. **Bilingual editing/post-editing** without verification — automated MT
   should be reviewed before publication.

## Limitations

1. **Hallucination on rare proper nouns**: Western names (Klein, Audrey,
   Bernadette) usually preserved, but uncommon proper nouns may hallucinate.
2. **Trained on web-novel corpus**: scientific, legal, or news domains
   may give suboptimal results.
3. **Long-context drift**: When translating a single long input (>200 chars
   in one go), proper names and consistent terminology may drift after
   several mentions in the same context (e.g., "Trương Vũ" → "Trương Huyền"
   → "Trương XX"). **Mitigation**: split long inputs by paragraph/sentence
   and translate each chunk independently. The HF Space demo applies this
   automatically.
4. **Output length asymptote**: outputs >1000 chars per chunk may degrade.
5. **Simplified Chinese only**: Traditional Chinese inputs untested and
   likely to degrade.

## Evaluation Methodology

Quality validation uses a **trio AI reviewer pattern** for cross-validated
human-style preference judgments without single-model bias.

### Reviewers

Three independent reviewer sessions, each using a different agent/runtime context:

- **Reviewer 1**: `gemini-3.1-pro` via Gemini CLI
- **Reviewer 2**: `gemini-3.5-flash` via Gemini CLI (different temperature)
- **Reviewer 3**: `gemini-3.5-flash` via Antigravity (AGY) agents

Each reviewer reads **one** review TSV in isolation — they cannot see other
reviewers' outputs.

### Scoring

Per row, per model:
- **Severity 0-3** scale (0 = OK / acceptable, 1 = minor error, 2 = moderate
  error, 3 = severe error — hallucination, truncation, or word salad)
- **Winner** pick: choose the best of 4 model outputs, or `tie` / `all_bad`
- **winner_reason** short text (model-specific failure modes or strengths)

### Aggregation

- **Pooled severity** = mean of all severity scores across reviewers (lower = better)
- **Winner aggregate** = vote count across 180 judgments (60 rows × 3 reviewers)
- **Trio consensus** = rows where all 3 reviewers agree on the same winner
  (highest-confidence signal)

### Test Sets

Two complementary evaluation sets covering web-novel translation diversity:

1. **Cross-novel paragraph** (60 rows, 20 short + 20 medium + 20 long buckets)
   — random paragraphs from two web-novels (Lovecraftian fantasy + sci-fi
   mecha), tests cross-domain + long-output handling.

2. **Xianxia in-distribution** (60 rows, 30 classical xianxia + 30 modern
   xianxia hybrid chapter excerpts) — tests xianxia genre quality and
   register polish (Hán Việt accuracy, tu tiên vocabulary, modern colloquial
   Vietnamese register).

### Anti-Bias Rules

To prevent single-reviewer drift:
- Each session opens **only one review file** (no cross-read)
- Anti-boilerplate rules enforced (no default severity=0, no default winner=tie)
- Reviewer-specific bias patterns identified post-hoc and weighted in interpretation

## Citation

```bibtex
@misc{hachimimt60-2026,
  author = {ngocdang83},
  title = {HachimiMT-60: Chinese-to-Vietnamese Web-Novel Translation},
  year = {2026},
  publisher = {Hugging Face},
  url = {https://huggingface.co/ngocdang83/HachimiMT-60-zh-vi}
}
```

## License

CC-BY-4.0 — free use with attribution. Training data includes Gemini API
teacher distillation; downstream users should verify current Gemini API
terms for derivative-work training.