17slever17

translate-gemma-4-sub-e2b

Deploy Dedicated

README

License: apache-2.0

Available formats

E4B:

Transformers: Translate Gemma 4 Sub E4B
GGUF: Translate Gemma 4 Sub E4B GGUF

E2B:

Transformers: Translate Gemma 4 Sub E2B
GGUF: Translate Gemma 4 Sub E2B GGUF

Intended use

The model is especially optimized for live and conversational speech, including streaming content, where meaning often depends on previous lines, speaker intent, tone, and incomplete context. It performs particularly well when previous source and translated segments are supplied and can follow additional user instructions that define the desired style, tone, or level of formality.

Translate Gemma 4 Sub retains the language coverage of the underlying Gemma 4 model. The most extensively trained languages are:

Tier 1: English, Russian, Spanish, German and Japanese;
Tier 2: French, Portuguese, Chinese, Dutch, Italian, and Korean.

Other languages supported by Gemma 4 may also work, but they have not undergone specific translation fine-tuning and have not been evaluated as extensively.

Benchmark results

All models in the following table were evaluated in GGUF Q4 format under the same generation and evaluation pipeline.

Higher is better for chrF++, BERTScore, COMET-DA, and COMETKiwi. Lower is better for MetricX Ref and MetricX QE.

Table with columns: Model, chrF++ ↑, BERTScore ↑, COMET-DA ↑, COMETKiwi ↑, MetricX-24 Ref ↓, MetricX-24 QE ↓
Model	chrF++ ↑	BERTScore ↑	COMET-DA ↑	COMETKiwi ↑	MetricX-24 Ref ↓	MetricX-24 QE ↓
Translate Gemma 4 Sub E4B Q4_K_XL	51.8218	0.867594	0.835393	0.745391	2.7074	2.8150
Translate Gemma 4 Sub E2B Q4_K_XL	49.1003	0.857130	0.822867	0.741567

The evaluation set was independent from the training data. Its reference translations were prepared from randomly selected stream segments that were not used during training.

MetricX-24 Ref by target language

Scores are grouped by target language. Lower is better.

Table with columns: Model, Russian ↓, English ↓, Japanese ↓, Spanish ↓, German ↓
Model	Russian ↓	English ↓	Japanese ↓	Spanish ↓	German ↓
Translate Gemma 4 Sub E4B Q4_K_XL	2.997	2.512	2.809	2.805	2.217
Translate Gemma 4 Sub E2B Q4_K_XL	3.100	2.573	3.138	2.901	2.370
Gemma 4 12B QAT Q4_K_XL

Prompt format

The model was trained with a system message that specifies the source language, target language, subtitle task, and style.

System message

text
TASK: Translate {source_language} subtitles into {target_language}.
RULES: Speakers name: ...; speaker gender: ...; other rules.
STYLE: friendly/official/neutral.
Translate only CURRENT_SOURCE. PREVIOUS_SOURCE and PREVIOUS_TRANSLATION are context only. Preserve meaning, tone, slang, profanity, uncertainty, repetitions and incomplete speech. Return only the final translation without labels or commentary.

User message

text
[PREVIOUS_SOURCE]
Previous source-language subtitles context.

[PREVIOUS_TRANSLATION]
Previous translated subtitles context.

[CURRENT_SOURCE]
The subtitle segment to translate.

Only CURRENT_SOURCE should be translated. PREVIOUS_SOURCE and PREVIOUS_TRANSLATION are context only.

The previous-context blocks may be omitted when no context is available:

text
[CURRENT_SOURCE]
The subtitle segment to translate.

Example

System message:

text
TASK: Translate English subtitles into Russian.
RULES: speaker gender: female.
STYLE: friendly.
Translate only CURRENT_SOURCE. PREVIOUS_SOURCE and PREVIOUS_TRANSLATION are context only. Preserve meaning, tone, slang, profanity, uncertainty, repetitions and incomplete speech. Return only the final translation without labels or commentary.

User message:

text
[PREVIOUS_SOURCE]
I thought you said you weren't coming.

[PREVIOUS_TRANSLATION]
Я думала, ты сказала, что не придёшь.

[CURRENT_SOURCE]
Yeah, well... I changed my mind.

Expected response:

text
Ну да... Я передумала.

FLORES-200 general translation benchmark

Evaluation was performed on FLORES-200 using MetricX-24 QE. Lower is better.

Table with columns: Model, Mean error ↓, Median ↓, P90 ↓
Model	Mean error ↓	Median ↓	P90 ↓
Gemma 4 12B QAT	1.8811	1.5391	3.5938
Translate Gemma 4 Sub E4B	2.0658	1.6523	4.0312
Gemma 4 E4B	2.0769	1.6875	4.1562
Translate Gemma 4 Sub E2B	2.2247	1.7930	4.4062

Despite being primarily optimized for contextual and conversational translation, Translate Gemma 4 Sub also slightly improved general translation quality over the corresponding base Gemma 4 models.

The following prompt was used for general translation:

text
TASK: Translate {source_language} into {target_language}.
Follow all demonstrations, glossary mappings, partial-translation constraints and formatting instructions in the user prompt.
Preserve meaning, names, numbers, terminology, register and document structure. Return only the requested final translation without commentary.

Usage with Transformers

python
import torch
from transformers import AutoModelForImageTextToText, AutoTokenizer

model_id = "17slever17/translate-gemma-4-sub-e2b"

tokenizer = AutoTokenizer.from_pretrained(model_id)

model = AutoModelForImageTextToText.from_pretrained(
    model_id,
    torch_dtype="auto",
    device_map="auto",
)

messages = [
    {
        "role": "system",
        "content": (
            "TASK: Translate English subtitles into Russian.\n"
            "STYLE: friendly.\n"
            "Translate only CURRENT_SOURCE. PREVIOUS_SOURCE and "
            "PREVIOUS_TRANSLATION are context only. Preserve meaning, tone, "
            "slang, profanity, uncertainty, repetitions and incomplete speech. "
            "Return only the final translation without labels or commentary."
        ),
    },
    {
        "role": "user",
        "content": (
            "[PREVIOUS_SOURCE]\n"
            "I thought you said you weren't coming.\n\n"
            "[PREVIOUS_TRANSLATION]\n"
            "Я думала, ты сказала, что не придёшь.\n\n"
            "[CURRENT_SOURCE]\n"
            "Yeah, well... I changed my mind."
        ),
    },
]

inputs = tokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
    return_tensors="pt",
    return_dict=True,
).to(model.device)

with torch.inference_mode():
    output = model.generate(
        **inputs,
        max_new_tokens=256,
        do_sample=False,
    )

generated_tokens = output[0, inputs["input_ids"].shape[1]:]
translation = tokenizer.decode(generated_tokens, skip_special_tokens=True)

print(translation)

Training overview

Translate Gemma 4 Sub E2B is based on google/gemma-4-E2B-it.

Training was performed with Unsloth using an optimized training run derived from the official Gemma 4 model.

Citation

bibtex
@misc{17slever17_translate_gemma_4_sub_2026,
  author       = {17slever17},
  title        = {Translate Gemma 4 Sub E2B},
  year         = {2026},
  publisher    = {Hugging Face},
  howpublished = {\url{https://huggingface.co/17slever17/translate-gemma-4-sub-e2b}}
}

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Container

Run this model inference with full control and performance in your environment.