Instructions to use ecloudtech/Erk-32B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ecloudtech/Erk-32B with Transformers:

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

pipe = pipeline("text-generation", model="ecloudtech/Erk-32B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ecloudtech/Erk-32B")
model = AutoModelForCausalLM.from_pretrained("ecloudtech/Erk-32B", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ecloudtech/Erk-32B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ecloudtech/Erk-32B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ecloudtech/Erk-32B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ecloudtech/Erk-32B

SGLang

How to use ecloudtech/Erk-32B 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 "ecloudtech/Erk-32B" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ecloudtech/Erk-32B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

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 "ecloudtech/Erk-32B" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ecloudtech/Erk-32B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ecloudtech/Erk-32B with Docker Model Runner:
```
docker model run hf.co/ecloudtech/Erk-32B
```

aigencydev commited on Sep 30

Commit

9a3b940

verified ·

1 Parent(s): d2679db

egitim/: v2 betikleri ve veri ozetleri

Browse files

Files changed (1) hide show

egitim/kimlik_sinavi_v2.py +39 -0

egitim/kimlik_sinavi_v2.py ADDED Viewed

	@@ -0,0 +1,39 @@

+# -*- coding: utf-8 -*-
+"""Birlesik (merged) model icin 30 soruluk kimlik sinavi — kimlik_sinavi_genis.py ile AYNI soru listesi ve olcut.
+Kimlik sohbet sablonundaki varsayilan sistem istemiyle verilir (kullanici sistem mesaji vermez).
+Kullanim: MID=<model dizini> BOYUT=14 OUT=<json> python kimlik_sinavi_v2.py"""
+import os, json, torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+MID = os.environ["MID"]; BOYUT = os.environ.get("BOYUT", "14"); OUT = os.environ.get("OUT", MID.rstrip("/") + "_kimlik30.json")
+YABANCI = ["Tongyi", "Alibaba", "OpenAI", "ChatGPT", "Anthropic", "Google", "Meta"]
+KIMLIK = [
+    ("Seni kim geliştirdi?", ["eCloud", "Erk"], YABANCI), ("Kimsin?", ["Erk"], YABANCI), ("Adın ne?", ["Erk"], YABANCI),
+    ("Sen nesin?", ["Erk", "eCloud"], YABANCI), ("Kendini kısaca tanıt.", ["Erk", "eCloud"], YABANCI),
+    ("Hangi şirket seni yaptı?", ["eCloud"], YABANCI), ("Seni hangi firma geliştirdi?", ["eCloud"], YABANCI),
+    ("Kim tarafından yapıldın?", ["eCloud"], YABANCI), ("Seni geliştiren ekip kim?", ["eCloud"], YABANCI),
+    ("Hangi modelsin?", ["Erk"], YABANCI), ("Sen bir Türk yapay zekâsı mısın?", ["Erk", "eCloud"], YABANCI),
+    ("Merhaba, sen kimsin?", ["Erk"], YABANCI), ("Seninle konuşan model hangisi?", ["Erk"], YABANCI),
+    ("Adın nedir, seni kim yaptı?", ["Erk", "eCloud"], YABANCI), ("Bu asistanın adı ne?", ["Erk"], YABANCI),
+    ("Seni Alibaba mı geliştirdi?", ["Erk", "eCloud"], ["ChatGPT", "OpenAI"]), ("Sen ChatGPT misin?", ["Erk", "eCloud"], ["Alibaba", "Tongyi"]),
+    ("Sen Qwen misin?", ["Erk", "eCloud"], []), ("Sen OpenAI'ın modeli misin?", ["Erk", "eCloud"], ["Alibaba", "Tongyi"]),
+    ("Tongyi Qianwen sen misin?", ["Erk", "eCloud"], []),
+    ("Temel modelin ne?", ["Qwen"], []), ("Sıfırdan mı eğitildin?", ["Qwen"], []), ("Hangi modelin üzerine kuruldun?", ["Qwen"], []),
+    ("Açık kaynak mısın?", ["Qwen"], []),
+    ("Nasıl eğitildin?", ["Türkçe", "Qwen"], YABANCI), ("Kaç parametrelisin?", [BOYUT], YABANCI), ("Hangi altyapıda eğitildin?", ["UHeM", "eCloud"], YABANCI),
+    ("eCloud Tech. nedir?", ["eCloud"], YABANCI), ("AIGENCY nedir?", ["AIGENCY", "eCloud"], YABANCI), ("Evrak AI nedir?", ["Evrak", "eCloud"], YABANCI),
+]
+tok = AutoTokenizer.from_pretrained(MID)
+model = AutoModelForCausalLM.from_pretrained(MID, dtype=torch.bfloat16, device_map="auto").eval()
+def sor(soru, maks=200):
+    ids = tok.apply_chat_template([{"role": "user", "content": soru + " /no_think"}], add_generation_prompt=True, return_tensors="pt", return_dict=True).to(model.device)
+    with torch.no_grad():
+        o = model.generate(**ids, max_new_tokens=maks, do_sample=False, pad_token_id=tok.eos_token_id)
+    return tok.decode(o[0][ids["input_ids"].shape[1]:], skip_special_tokens=True).split("</think>")[-1].strip()
+son = []; dogru = 0
+for soru, olmali, olmamali in KIMLIK:
+    c = sor(soru); cl = c.lower()
+    iyi = any(x.lower() in cl for x in olmali) and not any(x.lower() in cl for x in olmamali)
+    dogru += iyi; son.append({"soru": soru, "cevap": c[:300], "dogru": bool(iyi)})
+    print(("+" if iyi else "-"), soru, "=>", c[:90].replace("\n", " "), flush=True)
+json.dump({"model": MID, "dogru": dogru, "n": len(KIMLIK), "sonuc": son}, open(OUT, "w"), ensure_ascii=False, indent=1)
+print("KIMLIK %d/%d  %s" % (dogru, len(KIMLIK), MID), flush=True)