Instructions to use icefog72/Ice0.7-29.09-RP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use icefog72/Ice0.7-29.09-RP with Transformers:

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

pipe = pipeline("text-generation", model="icefog72/Ice0.7-29.09-RP")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("icefog72/Ice0.7-29.09-RP")
model = AutoModelForCausalLM.from_pretrained("icefog72/Ice0.7-29.09-RP", 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 icefog72/Ice0.7-29.09-RP with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "icefog72/Ice0.7-29.09-RP"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "icefog72/Ice0.7-29.09-RP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/icefog72/Ice0.7-29.09-RP

SGLang

How to use icefog72/Ice0.7-29.09-RP 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 "icefog72/Ice0.7-29.09-RP" \
    --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": "icefog72/Ice0.7-29.09-RP",
		"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 "icefog72/Ice0.7-29.09-RP" \
        --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": "icefog72/Ice0.7-29.09-RP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use icefog72/Ice0.7-29.09-RP with Docker Model Runner:
```
docker model run hf.co/icefog72/Ice0.7-29.09-RP
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Ice0.7-29.09-RP

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the SLERP merge method.

Models Merged

The following models were included in the merge:

F:\FModels\Ice0.5-28.09-RP
E:\FModels\Ice0.6-29.09-RP

Configuration

The following YAML configuration was used to produce this model:

slices:
  - sources:
      - model: E:\FModels\Ice0.6-29.09-RP
        layer_range: [0, 32]
      - model: F:\FModels\Ice0.5-28.09-RP
        layer_range: [0, 32]

merge_method: slerp
base_model: F:\FModels\Ice0.5-28.09-RP
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5 # fallback for rest of tensors
dtype: bfloat16

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	21.47
IFEval (0-Shot)	51.76
BBH (3-Shot)	30.73
MATH Lvl 5 (4-Shot)	6.19
GPQA (0-shot)	5.03
MuSR (0-shot)	11.51
MMLU-PRO (5-shot)	23.63

Downloads last month: 14

Safetensors

Model size

7B params

Tensor type

BF16

Model tree for icefog72/Ice0.7-29.09-RP

Quantizations

2 models

Evaluation results

strict accuracy on IFEval (0-Shot)
Open LLM Leaderboard

51.760
normalized accuracy on BBH (3-Shot)
Open LLM Leaderboard

30.730
exact match on MATH Lvl 5 (4-Shot)
Open LLM Leaderboard

6.190
acc_norm on GPQA (0-shot)
Open LLM Leaderboard

5.030
acc_norm on MuSR (0-shot)
Open LLM Leaderboard

11.510
accuracy on MMLU-PRO (5-shot)
test set Open LLM Leaderboard

23.630