Instructions to use tcclaviger/Laguna-S-2.1-RFA_L with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use tcclaviger/Laguna-S-2.1-RFA_L with Transformers:

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

pipe = pipeline("text-generation", model="tcclaviger/Laguna-S-2.1-RFA_L", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("tcclaviger/Laguna-S-2.1-RFA_L", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("tcclaviger/Laguna-S-2.1-RFA_L", trust_remote_code=True, 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 tcclaviger/Laguna-S-2.1-RFA_L with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "tcclaviger/Laguna-S-2.1-RFA_L"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "tcclaviger/Laguna-S-2.1-RFA_L",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/tcclaviger/Laguna-S-2.1-RFA_L

SGLang

How to use tcclaviger/Laguna-S-2.1-RFA_L 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 "tcclaviger/Laguna-S-2.1-RFA_L" \
    --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": "tcclaviger/Laguna-S-2.1-RFA_L",
		"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 "tcclaviger/Laguna-S-2.1-RFA_L" \
        --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": "tcclaviger/Laguna-S-2.1-RFA_L",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use tcclaviger/Laguna-S-2.1-RFA_L with Docker Model Runner:
```
docker model run hf.co/tcclaviger/Laguna-S-2.1-RFA_L
```

tcclaviger commited on 4 days ago

Commit

219e948

verified ·

1 Parent(s): 24a49fa

Add eval results notice to model card

Browse files

Files changed (1) hide show

README.md +21 -0

README.md CHANGED Viewed

@@ -34,6 +34,27 @@ base_model_relation: quantized
 > exactly as documented in the vLLM section below — the RFA quant does not touch the
 > attention path, so the DFlash draft head runs unchanged.
 ---
 <p align="center">

 > exactly as documented in the vLLM section below — the RFA quant does not touch the
 > attention path, so the DFlash draft head runs unchanged.
+> [!NOTE]
+> ## Evaluation results (this RFA quant)
+>
+> | Eval | Result |
+> |---|---|
+> | Throughput | 80.6 tok/s out @ conc 1 (DFlash accept ~2.1); 615.7 tok/s out @ conc 50, ISL 512 |
+> | WikiText-2 PPL | 8.109 ± 0.043 (n_ctx 2048, llama.cpp-compatible) |
+> | Loglikelihood acc | arc_challenge 0.503 / arc_easy 0.769 / winogrande 0.658 / hellaswag 0.814 (acc_norm) |
+> | tool-eval-bench (no-think) | 89/100 (full 69 scenarios) |
+> | GSM8K / MMLU / IFEval | 90% / 60% / 95% (prompt-level) |
+> | Long-context code recall | 0.980 overall (py 1.00 / js 0.93 / rs 0.99 / cpp 1.00); tool pass 93.5% |
+> | Hard Mode agentic (thinking-ON) | 73/100 — 10 pass / 2 partial / 3 fail |
+>
+> **Test harnesses:** `vllm bench serve` (random dataset, saturation sweep) ·
+> llama.cpp-compatible WikiText-2 perplexity · lm-evaluation-harness (loglikelihood,
+> local-completions) · tool-eval-bench v2 (69 scenarios + GSM8K/MMLU/IFEval + 15
+> Hard Mode scenarios) · codeneedle (positional recall, 4 corpora + tool pass).
+>
+> **Hardware:** `tcclaviger/vllm:latest`, 4× AMD AI PRO R9700 GPUs (TP4), Ryzen 9
+> 9950X, 256 GB DDR5-6000. (Not yet fully tuned for throughput.)
 ---
 <p align="center">