Instructions to use ICONNAI/ICONN-e1-Beta with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ICONNAI/ICONN-e1-Beta with Transformers:

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

pipe = pipeline("text-generation", model="ICONNAI/ICONN-e1-Beta")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ICONNAI/ICONN-e1-Beta")
model = AutoModelForCausalLM.from_pretrained("ICONNAI/ICONN-e1-Beta", 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ICONNAI/ICONN-e1-Beta with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ICONNAI/ICONN-e1-Beta"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ICONNAI/ICONN-e1-Beta",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/ICONNAI/ICONN-e1-Beta

SGLang

How to use ICONNAI/ICONN-e1-Beta 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 "ICONNAI/ICONN-e1-Beta" \
    --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": "ICONNAI/ICONN-e1-Beta",
		"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 "ICONNAI/ICONN-e1-Beta" \
        --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": "ICONNAI/ICONN-e1-Beta",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use ICONNAI/ICONN-e1-Beta with Docker Model Runner:
```
docker model run hf.co/ICONNAI/ICONN-e1-Beta
```

Enderchef commited on Jun 17, 2025

Commit

24ff901

verified ·

1 Parent(s): 54c6f28

Add mergekit_moe_config.yml

Browse files

Files changed (1) hide show

mergekit_moe_config.yml +104 -0

mergekit_moe_config.yml ADDED Viewed

	@@ -0,0 +1,104 @@

+output_model: ./ICONN-1
+dtype: bfloat16
+moe_type: task
+tokenizer_source: models/arcee-blitz
+base_model:
+  model: models/arcee-blitz
+  source_model: arcee-ai/Arcee-Blitz  # base model source repo
+experts:
+  - name: CodeCONN
+    model: models/devstral
+    source_model: mistralai/Devstral-Small-2505
+    positive_prompts:
+      - "python"
+      - "java"
+      - "c++"
+      - "programming"
+      - "//"
+      - "code"
+      - "syntax"
+      - "pandas"
+      - "sql"
+      - "import"
+      - "data_cleaning"
+      - "script"
+      - "bug"
+      - "fix"
+      - "loop"
+      - "json"
+      - "csv"
+      - "binary"
+      - "compiler"
+      - "function"
+      - "class"
+  - name: ChatCONN
+    model: models/dolphin3
+    source_model: cognitivecomputations/Dolphin3.0-Mistral-24B
+    positive_prompts:
+      - "hi"
+      - "hello"
+      - "how are you"
+      - "what's up"
+      - "can you help"
+      - "tell me a joke"
+      - "write me"
+      - "explain"
+      - "talk to me"
+      - "describe"
+      - "friendly"
+      - "respond"
+      - "short answer"
+      - "summarize"
+      - "conversation"
+      - "chat"
+      - "help me"
+  - name: ReasonCONN
+    model: models/dolphin3-r1
+    source_model: cognitivecomputations/Dolphin3.0-R1-Mistral-24B
+    positive_prompts:
+      - "think"
+      - "logic"
+      - "reason"
+      - "math"
+      - "why"
+      - "how"
+      - "deduce"
+      - "explain in detail"
+      - "analyze"
+      - "IQ"
+      - "SAT"
+      - "GRE"
+      - "proof"
+      - "science"
+      - "physics"
+      - "biology"
+      - "deepen"
+      - "calculus"
+      - "infer"
+  - name: GeoCONN
+    model: models/deephermes
+    source_model: NousResearch/DeepHermes-3-Mistral-24B-Preview
+    positive_prompts:
+      - "geography"
+      - "continent"
+      - "country"
+      - "capital"
+      - "map"
+      - "longitude"
+      - "latitude"
+      - "earth"
+      - "region"
+      - "nation"
+      - "location"
+      - "mountain"
+      - "river"
+      - "climate"
+      - "ocean"
+      - "desert"
+      - "geological"
+      - "atlas"