Instructions to use evilfreelancer/ruGPT3XL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use evilfreelancer/ruGPT3XL with Transformers:

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

pipe = pipeline("text-generation", model="evilfreelancer/ruGPT3XL", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("evilfreelancer/ruGPT3XL", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use evilfreelancer/ruGPT3XL with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "evilfreelancer/ruGPT3XL"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "evilfreelancer/ruGPT3XL",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/evilfreelancer/ruGPT3XL

SGLang

How to use evilfreelancer/ruGPT3XL 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 "evilfreelancer/ruGPT3XL" \
    --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": "evilfreelancer/ruGPT3XL",
		"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 "evilfreelancer/ruGPT3XL" \
        --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": "evilfreelancer/ruGPT3XL",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use evilfreelancer/ruGPT3XL with Docker Model Runner:
```
docker model run hf.co/evilfreelancer/ruGPT3XL
```

Pavel Rykov commited on Mar 27

Commit

d01e7f2

2 Parent(s): 35af8a3 00c8d90

Merge branch 'main' of hf.co:evilfreelancer/ruGPT3XL

Browse files

Files changed (1) hide show

README.md +3 -3

README.md CHANGED Viewed

@@ -21,7 +21,7 @@ into a native HuggingFace `transformers` format.
 This is a **base (pretrained) model**, not instruction-tuned. It performs text completion
 and can be fine-tuned for downstream tasks.
-Details in "[A family of pretrained transformer language models for Russian](https://scholar.google.com/citations?view_op=view_citation&hl=en&user=yPayeJIAAAAJ&citation_for_view=yPayeJIAAAAJ:Se3iqnhoufwC)" paper.
 ## Model Details
@@ -232,7 +232,7 @@ The model implements a custom `RuGPT3XLForCausalLM` class (loaded via `trust_rem
 ```
 RuGPT3XLForCausalLM
   ├── model (RuGPT3XLModel)
-  │     ├── embed_tokens      (Embedding: 50264 x 2048)
   │     ├── embed_positions    (Embedding: 2048 x 2048)
   │     ├── embed_dropout      (Dropout: 0.1)
   │     ├── layers (x24)       (RuGPT3XLDecoderLayer)
@@ -244,7 +244,7 @@ RuGPT3XLForCausalLM
   │     │     │     ├── o_proj             (Linear: 2048 -> 2048)
   │     │     │     ├── attn_dropout       (Dropout: 0.1)
   │     │     │     └── resid_dropout      (Dropout: 0.1)
-  │     │     ├── post_attention_layernorm  (LayerNorm: 2048)
   │     │     └── mlp                      (RuGPT3XMLP)
   │     │           ├── up_proj            (Linear: 2048 -> 8192)
   │     │           ├── down_proj          (Linear: 8192 -> 2048)

 This is a **base (pretrained) model**, not instruction-tuned. It performs text completion
 and can be fine-tuned for downstream tasks.
+See more in "[A family of pretrained transformer language models for Russian](https://scholar.google.com/citations?view_op=view_citation&hl=en&user=yPayeJIAAAAJ&citation_for_view=yPayeJIAAAAJ:Se3iqnhoufwC)" paper.
 ## Model Details
 ```
 RuGPT3XLForCausalLM
   ├── model (RuGPT3XLModel)
+  │     ├── embed_tokens       (Embedding: 50264 x 2048)
   │     ├── embed_positions    (Embedding: 2048 x 2048)
   │     ├── embed_dropout      (Dropout: 0.1)
   │     ├── layers (x24)       (RuGPT3XLDecoderLayer)
   │     │     │     ├── o_proj             (Linear: 2048 -> 2048)
   │     │     │     ├── attn_dropout       (Dropout: 0.1)
   │     │     │     └── resid_dropout      (Dropout: 0.1)
+  │     │     ├── post_attention_layernorm (LayerNorm: 2048)
   │     │     └── mlp                      (RuGPT3XMLP)
   │     │           ├── up_proj            (Linear: 2048 -> 8192)
   │     │           ├── down_proj          (Linear: 8192 -> 2048)