Instructions to use EvanOLeary/laguna-xs2-dense-k8-cuda-sft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use EvanOLeary/laguna-xs2-dense-k8-cuda-sft with Transformers:

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

pipe = pipeline("text-generation", model="EvanOLeary/laguna-xs2-dense-k8-cuda-sft", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("EvanOLeary/laguna-xs2-dense-k8-cuda-sft", trust_remote_code=True, device_map="auto")

Kernels

How to use EvanOLeary/laguna-xs2-dense-k8-cuda-sft with Kernels:

# !pip install kernels

from kernels import get_kernel

# a version (or an explicit revision) is required; see the "Files and versions" tab for the available ones
kernel = get_kernel("EvanOLeary/laguna-xs2-dense-k8-cuda-sft", version=1)

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use EvanOLeary/laguna-xs2-dense-k8-cuda-sft with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "EvanOLeary/laguna-xs2-dense-k8-cuda-sft"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "EvanOLeary/laguna-xs2-dense-k8-cuda-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/EvanOLeary/laguna-xs2-dense-k8-cuda-sft

SGLang

How to use EvanOLeary/laguna-xs2-dense-k8-cuda-sft 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 "EvanOLeary/laguna-xs2-dense-k8-cuda-sft" \
    --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": "EvanOLeary/laguna-xs2-dense-k8-cuda-sft",
		"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 "EvanOLeary/laguna-xs2-dense-k8-cuda-sft" \
        --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": "EvanOLeary/laguna-xs2-dense-k8-cuda-sft",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use EvanOLeary/laguna-xs2-dense-k8-cuda-sft with Docker Model Runner:
```
docker model run hf.co/EvanOLeary/laguna-xs2-dense-k8-cuda-sft
```

EvanOLeary commited on May 30

Commit

7194328

verified ·

1 Parent(s): 2afa69c

Upload README.md with huggingface_hub

Browse files

Files changed (1) hide show

README.md +24 -0

README.md CHANGED Viewed

@@ -54,6 +54,30 @@ row = next(iter(ds))
 prompt, target, ok = row["PyTorch_Code_Module"], row["CUDA_Code"], row["Correct"]
 ```
 ## KernelBench-Lite baseline (this SFT model)
 generate → compile (`load_inline`) → correctness vs PyTorch eager → speedup:
 - **ReLU:** compiles + correct ✅, **0.93×** vs eager (best-of-4). It produces real, working CUDA.

 prompt, target, ok = row["PyTorch_Code_Module"], row["CUDA_Code"], row["Correct"]
 ```
+## Training overview — data, steps, DSLs (graphs)
+![overview](https://huggingface.co/EvanOLeary/laguna-xs2-dense-k8-cuda-sft/resolve/main/training_overview.png)
+### Stages & steps
+| Stage | Steps | Tokens | Data | Trainable |
+|---|---|---|---|---|
+| 0 Warm-start (DO-ACP) | — | — | calibration only | — |
+| 1 Recon V1 (Python) | 2000 | ~8.2M | OpenCodeInstruct | routed_dense |
+| 1 Recon V2 (kernel) | 2000 | ~8.2M | **KernelBook40/OpenCode30/CUDA20/traces10** | routed_dense |
+| 2 SFT (CUDA) | 400 | ~3.5M | SakanaAI CUDA (correct only) | routed_dense+lm_head+norms |
+### Data distribution (reconstruction V2 mixture)
+KernelBook (Triton) 40% · OpenCodeInstruct (Python) 30% · **SakanaAI CUDA-C++ 20%** · Triton traces 10% — ≈ **50% kernel / 30% Python / 20% CUDA-C++**.
+### DSLs
+- **CUDA** (SFT-trained) — stronger; simple ops compile+correct at pass@k.
+- **Triton** (pretrain-only) — emits valid `@triton.jit` with a Triton prompt, but buggier idioms.
+### Reconstruction curves (V2 kernel-mix, 2000 steps)
+![v2 recon](https://huggingface.co/EvanOLeary/laguna-xs2-dense-k8-kernelmix/resolve/main/v2_training.png)
+### SFT curve (CUDA, 400 steps, CE 0.68→0.32)
+![sft](https://huggingface.co/EvanOLeary/laguna-xs2-dense-k8-cuda-sft/resolve/main/sft_curve.png)
 ## KernelBench-Lite baseline (this SFT model)
 generate → compile (`load_inline`) → correctness vs PyTorch eager → speedup:
 - **ReLU:** compiles + correct ✅, **0.93×** vs eager (best-of-4). It produces real, working CUDA.