Instructions to use IFM/K2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use IFM/K2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="IFM/K2")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("IFM/K2") model = AutoModelForCausalLM.from_pretrained("IFM/K2", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use IFM/K2 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "IFM/K2" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "IFM/K2", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/IFM/K2
- SGLang
How to use IFM/K2 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 "IFM/K2" \ --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": "IFM/K2", "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 "IFM/K2" \ --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": "IFM/K2", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use IFM/K2 with Docker Model Runner:
docker model run hf.co/IFM/K2
Update README.md
Browse files
README.md
CHANGED
|
@@ -96,7 +96,21 @@ We provide step-by-step reproducation tutorials for tech enthusiasts, AI practit
|
|
| 96 |
## LLM360 Developer Suite
|
| 97 |
We provide step-by-step finetuning tutorials for tech enthusiasts, AI practitioners and academic or industry researchers here [llm360.ai/pretraining].
|
| 98 |
|
| 99 |
-
#
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 100 |
|
| 101 |
|
| 102 |
|
|
|
|
| 96 |
## LLM360 Developer Suite
|
| 97 |
We provide step-by-step finetuning tutorials for tech enthusiasts, AI practitioners and academic or industry researchers here [llm360.ai/pretraining].
|
| 98 |
|
| 99 |
+
# Loading K2
|
| 100 |
+
```python
|
| 101 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 102 |
+
|
| 103 |
+
tokenizer = AutoTokenizer.from_pretrained("LLM360/K2")
|
| 104 |
+
model = AutoModelForCausalLM.from_pretrained("LLM360/K2")
|
| 105 |
+
|
| 106 |
+
prompt = 'int add(int x, int y) {'
|
| 107 |
+
|
| 108 |
+
input_ids = tokenizer(prompt, return_tensors="pt").input_ids
|
| 109 |
+
gen_tokens = model.generate(input_ids, do_sample=True, max_length=128)
|
| 110 |
+
|
| 111 |
+
print("-"*20 + "Output for model" + 20 * '-')
|
| 112 |
+
print(tokenizer.batch_decode(gen_tokens)[0])
|
| 113 |
+
```
|
| 114 |
|
| 115 |
|
| 116 |
|