Instructions to use oddadmix/Emhotob-10M-Egyptian-English-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use oddadmix/Emhotob-10M-Egyptian-English-v2 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="oddadmix/Emhotob-10M-Egyptian-English-v2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("oddadmix/Emhotob-10M-Egyptian-English-v2")
model = AutoModelForCausalLM.from_pretrained("oddadmix/Emhotob-10M-Egyptian-English-v2", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use oddadmix/Emhotob-10M-Egyptian-English-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "oddadmix/Emhotob-10M-Egyptian-English-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "oddadmix/Emhotob-10M-Egyptian-English-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/oddadmix/Emhotob-10M-Egyptian-English-v2

SGLang

How to use oddadmix/Emhotob-10M-Egyptian-English-v2 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "oddadmix/Emhotob-10M-Egyptian-English-v2" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "oddadmix/Emhotob-10M-Egyptian-English-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "oddadmix/Emhotob-10M-Egyptian-English-v2" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "oddadmix/Emhotob-10M-Egyptian-English-v2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use oddadmix/Emhotob-10M-Egyptian-English-v2 with Docker Model Runner:
```
docker model run hf.co/oddadmix/Emhotob-10M-Egyptian-English-v2
```

oddadmix commited on Jul 19

Commit

4f789c2

verified ·

1 Parent(s): e7d1ec5

Add out-of-domain English->Egyptian leaderboard eval

Browse files

Files changed (1) hide show

README.md +10 -8

README.md CHANGED Viewed

@@ -106,13 +106,13 @@ def translate(text, system=SYSTEM):
 - **Eval split:** 3,000 deterministic held-out pairs (`seed=42`), scored both directions.
 <!-- BEGIN: leaderboard-en2egy-eval -->
-## Out-of-domain evaluation — Egyptian leaderboard set
 The results above are **in-domain**: a held-out split of the same corpus this model was
-trained on. The numbers below are **out-of-domain** — the same model scored on
-[`oddadmix/egyptian_translation_eval_nourhan`](https://huggingface.co/datasets/oddadmix/egyptian_translation_eval_nourhan) (319 English→Egyptian pairs,
-a different annotator with different orthographic conventions), the eval set used by the
-Egyptian translation leaderboard.
 Expect these to be substantially lower than the in-domain scores. That gap is the
 generalization penalty, not a regression — both numbers are real, they measure different things.
@@ -149,9 +149,11 @@ translation that picks a different valid word is penalized — e.g. `فريش` v
 good Egyptian; only one matches the reference. **chrF and METEOR track perceived quality more
 closely here.**
-These are small models — 5M to 50M parameters, orders of magnitude below the large systems that
-top this leaderboard. The result of interest is the **scaling curve and per-parameter
-efficiency**, not absolute rank against models 100–1000× the size.
 <!-- END: leaderboard-en2egy-eval -->
 ## Limitations

 - **Eval split:** 3,000 deterministic held-out pairs (`seed=42`), scored both directions.
 <!-- BEGIN: leaderboard-en2egy-eval -->
+## Out-of-domain evaluation — Egyptian Arabic Translation Benchmark
 The results above are **in-domain**: a held-out split of the same corpus this model was
+trained on. The numbers below are **out-of-domain** — the same model scored on the
+[Egyptian Arabic Translation Benchmark](https://huggingface.co/datasets/oddadmix/egyptian-arabic-translation-benchmark)
+(`oddadmix/egyptian-arabic-translation-benchmark`, 319 English→Egyptian pairs written by a different annotator with different
+orthographic conventions).
 Expect these to be substantially lower than the in-domain scores. That gap is the
 generalization penalty, not a regression — both numbers are real, they measure different things.
 good Egyptian; only one matches the reference. **chrF and METEOR track perceived quality more
 closely here.**
+At 319 rows, differences of roughly 1–2 BLEU between adjacent rungs are within noise.
+These are small models — 5M to 50M parameters, orders of magnitude below the large systems
+typically evaluated on this benchmark. The result of interest is the **scaling curve and
+per-parameter efficiency**, not absolute rank against models 100–1000× the size.
 <!-- END: leaderboard-en2egy-eval -->
 ## Limitations