Text Generation
Transformers
Safetensors
slonik
code
postgresql
software-engineering
reinforcement-learning
Instructions to use dglhf/Slonik with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use dglhf/Slonik with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="dglhf/Slonik")# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("dglhf/Slonik", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use dglhf/Slonik with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "dglhf/Slonik" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "dglhf/Slonik", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/dglhf/Slonik
- SGLang
How to use dglhf/Slonik 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 "dglhf/Slonik" \ --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": "dglhf/Slonik", "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 "dglhf/Slonik" \ --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": "dglhf/Slonik", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use dglhf/Slonik with Docker Model Runner:
docker model run hf.co/dglhf/Slonik
Link Postgres Professional and simplify card header
Browse files
README.md
CHANGED
|
@@ -25,7 +25,7 @@ November 2026
|
|
| 25 |
|
| 26 |
Database engines are critical infrastructure, yet their development remains an underexplored setting for coding agents. A PostgreSQL patch must connect systems code with behavior exposed through SQL; successful compilation alone does not establish correctness. Existing database environments often study [tuning](https://vldb.org/cidrdb/2023/database-gyms.html) or [SQL debugging](https://arxiv.org/abs/2506.18951), while engine-development tasks require changes to the database implementation itself.
|
| 27 |
|
| 28 |
-
Slonik comprises two language models for agentic coding (Slonik-27B, Slonik-9B), developed at **Postgres Professional**. We use PostgreSQL engine development to train the models, then test whether their gains extend to general software engineering tasks.
|
| 29 |
|
| 30 |
## Model Details
|
| 31 |
|
|
@@ -104,5 +104,5 @@ PG17 informed model selection; several transfer suites were inspected repeatedly
|
|
| 104 |
## Authors
|
| 105 |
|
| 106 |
**Dmtriy Gilemkhanov, Vladlen Popolitov, Saveliy Baturin, Dmitrii Tarasov, Oleg Bartunov**
|
| 107 |
-
Postgres Professional
|
| 108 |
|
|
|
|
| 25 |
|
| 26 |
Database engines are critical infrastructure, yet their development remains an underexplored setting for coding agents. A PostgreSQL patch must connect systems code with behavior exposed through SQL; successful compilation alone does not establish correctness. Existing database environments often study [tuning](https://vldb.org/cidrdb/2023/database-gyms.html) or [SQL debugging](https://arxiv.org/abs/2506.18951), while engine-development tasks require changes to the database implementation itself.
|
| 27 |
|
| 28 |
+
Slonik comprises two language models for agentic coding (Slonik-27B, Slonik-9B), developed at **[Postgres Professional](https://postgrespro.com/)**. We use PostgreSQL engine development to train the models, then test whether their gains extend to general software engineering tasks.
|
| 29 |
|
| 30 |
## Model Details
|
| 31 |
|
|
|
|
| 104 |
## Authors
|
| 105 |
|
| 106 |
**Dmtriy Gilemkhanov, Vladlen Popolitov, Saveliy Baturin, Dmitrii Tarasov, Oleg Bartunov**
|
| 107 |
+
[Postgres Professional](https://postgrespro.com/)
|
| 108 |
|