Instructions to use hellohle/imlong with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hellohle/imlong with Transformers:

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("hellohle/imlong")
model = AutoModelForCausalLM.from_pretrained("hellohle/imlong", 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 hellohle/imlong with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/hellohle/imlong

SGLang

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

Docker Model Runner
How to use hellohle/imlong with Docker Model Runner:
```
docker model run hf.co/hellohle/imlong
```

hellohle commited on Mar 16

Commit

2926dcc

verified ·

1 Parent(s): b84ab16

Update config.json

Browse files

Files changed (1) hide show

config.json +30 -20

config.json CHANGED Viewed

@@ -1,29 +1,39 @@
 {
-  "architectures": ["LlamaForCausalLM"],
-  "model_type": "llama",
-  "hidden_size": 3584,
-  "intermediate_size": 14336,
-  "num_attention_heads": 28,
-  "num_key_value_heads": 28,
-  "num_hidden_layers": 28,
-  "vocab_size": 152064,
-  "max_position_embeddings": 16384,
   "hidden_act": "silu",
   "initializer_range": 0.02,
-  "rms_norm_eps": 1e-6,
-  "tie_word_embeddings": false,
-  "bos_token_id": 32013,
-  "eos_token_id": 32021,
-  "rope_theta": 100000,
-  "use_cache": true
 }

 {
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 32013,
+  "dtype": "float16",
+  "eos_token_id": 32021,
+  "head_dim": 128,
   "hidden_act": "silu",
+  "hidden_size": 4096,
   "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 16384,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 32,
+  "pad_token_id": null,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "factor": 4.0,
+    "rope_theta": 100000,
+    "rope_type": "linear",
+    "type": "linear"
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.0.0",
+  "use_cache": true,
+  "vocab_size": 32256
 }