Instructions to use tiny-random/glm-5.2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tiny-random/glm-5.2 with Transformers:

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

pipe = pipeline("text-generation", model="tiny-random/glm-5.2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tiny-random/glm-5.2")
model = AutoModelForCausalLM.from_pretrained("tiny-random/glm-5.2", 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 tiny-random/glm-5.2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tiny-random/glm-5.2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tiny-random/glm-5.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tiny-random/glm-5.2

SGLang

How to use tiny-random/glm-5.2 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 "tiny-random/glm-5.2" \
    --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": "tiny-random/glm-5.2",
		"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 "tiny-random/glm-5.2" \
        --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": "tiny-random/glm-5.2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tiny-random/glm-5.2 with Docker Model Runner:
```
docker model run hf.co/tiny-random/glm-5.2
```

yujiepan commited on 30 days ago

Commit

35f2a8a

verified ·

1 Parent(s): cfdcfbb

Upload folder using huggingface_hub

Browse files

Files changed (3) hide show

README.md +6 -6
config.json +1 -1
model.safetensors +2 -2

README.md CHANGED Viewed

@@ -8,7 +8,7 @@ This tiny model is intended for debugging. It is randomly initialized using the
 | File path | Size |
 |------|------|
-| model.safetensors | 26.2MB |
 ### Example usage:
@@ -98,7 +98,7 @@ config_json.update({
     "first_k_dense_replace": 1,
     "mlp_layer_types": ['dense'] + ['sparse'] * 3,
     "hidden_size": 8,
-    "index_n_heads": 4,
     "indexer_types": ['full'] + ['shared'] * 3,
     "index_topk_pattern": ['F'] + ['S'] * 3,
     "intermediate_size": 32,
@@ -181,10 +181,10 @@ GlmMoeDsaForCausalLM(
           (kv_b_proj): Linear(in_features=512, out_features=3584, bias=False)
           (o_proj): Linear(in_features=2048, out_features=8, bias=False)
           (indexer): GlmMoeDsaIndexer(
-            (wq_b): Linear(in_features=32, out_features=512, bias=False)
             (wk): Linear(in_features=8, out_features=128, bias=False)
             (k_norm): LayerNorm((128,), eps=1e-06, elementwise_affine=True)
-            (weights_proj): Linear(in_features=8, out_features=4, bias=False)
           )
         )
         (mlp): GlmMoeDsaMLP(
@@ -239,10 +239,10 @@ GlmMoeDsaForCausalLM(
           (kv_b_proj): Linear(in_features=512, out_features=3584, bias=False)
           (o_proj): Linear(in_features=2048, out_features=8, bias=False)
           (indexer): GlmMoeDsaIndexer(
-            (wq_b): Linear(in_features=32, out_features=512, bias=False)
             (wk): Linear(in_features=8, out_features=128, bias=False)
             (k_norm): LayerNorm((128,), eps=1e-06, elementwise_affine=True)
-            (weights_proj): Linear(in_features=8, out_features=4, bias=False)
           )
         )
         (mlp): GlmMoeDsaMoE(

 | File path | Size |
 |------|------|
+| model.safetensors | 26.7MB |
 ### Example usage:
     "first_k_dense_replace": 1,
     "mlp_layer_types": ['dense'] + ['sparse'] * 3,
     "hidden_size": 8,
+    "index_n_heads": 32,
     "indexer_types": ['full'] + ['shared'] * 3,
     "index_topk_pattern": ['F'] + ['S'] * 3,
     "intermediate_size": 32,
           (kv_b_proj): Linear(in_features=512, out_features=3584, bias=False)
           (o_proj): Linear(in_features=2048, out_features=8, bias=False)
           (indexer): GlmMoeDsaIndexer(
+            (wq_b): Linear(in_features=32, out_features=4096, bias=False)
             (wk): Linear(in_features=8, out_features=128, bias=False)
             (k_norm): LayerNorm((128,), eps=1e-06, elementwise_affine=True)
+            (weights_proj): Linear(in_features=8, out_features=32, bias=False)
           )
         )
         (mlp): GlmMoeDsaMLP(
           (kv_b_proj): Linear(in_features=512, out_features=3584, bias=False)
           (o_proj): Linear(in_features=2048, out_features=8, bias=False)
           (indexer): GlmMoeDsaIndexer(
+            (wq_b): Linear(in_features=32, out_features=4096, bias=False)
             (wk): Linear(in_features=8, out_features=128, bias=False)
             (k_norm): LayerNorm((128,), eps=1e-06, elementwise_affine=True)
+            (weights_proj): Linear(in_features=8, out_features=32, bias=False)
           )
         )
         (mlp): GlmMoeDsaMoE(

config.json CHANGED Viewed

@@ -17,7 +17,7 @@
   "hidden_act": "silu",
   "hidden_size": 8,
   "index_head_dim": 128,
-  "index_n_heads": 4,
   "index_share_for_mtp_iteration": true,
   "index_skip_topk_offset": 3,
   "index_topk": 2048,

   "hidden_act": "silu",
   "hidden_size": 8,
   "index_head_dim": 128,
+  "index_n_heads": 32,
   "index_share_for_mtp_iteration": true,
   "index_skip_topk_offset": 3,
   "index_topk": 2048,

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:151d405ef695562d64c615d24d4f0cbf0aec60d7b563c545ceb41484959a0744
-size 26216056

 version https://git-lfs.github.com/spec/v1
+oid sha256:c380b6f0d6d1560368508cc5b64580e316754aa935aadf1b210de43b8650c86a
+size 26675712