Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →mkurman/sglang
Structured Generation Language for LLM serving. RadixAttention prefix caching, constrained decoding (JSON, grammar), OpenAI-compatible API, and multi-turn optimization. Fast inference with structured output guarantees.
Passed 21 of 22 test cases with the skill, and 18 of 22 without it.
3 fixed18 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 3 test cases the skill fixed. Same model, same prompt, with and without the skill.
When creating an OpenAI client in Python to connect to a locally launched SGLang server that requires no authentication token, what string value should be passed for `api_key`? I am considering passing `None` or omitting the argument.
| Checked | Without skill | With skill |
|---|---|---|
| passes string literal none as the api_key | Missed | Passed |
The instructions your model reads when the skill loads.
SGLang is a fast LLM inference and serving framework with structured generation (constrained decoding), RadixAttention for prefix caching, and an OpenAI-compatible API. Particularly strong for reasoning models (DeepSeek R1, QwQ) and guided generation with JSON schema, grammar, and regex constraints.
uv pip install sglang[all]
import sglang as sgl
@sgl.function
def multi_turn(s, question):
s += sgl.system("You are a helpful assistant.")
s += sgl.user(question)
s += sgl.assistant()
state = multi_turn.run(question="What is the derivative of x^2?")
print(state["answer"])
python -m sglang.launch_server --model-path Qwen/Qwen2.5-1.5B-Instruct --port 30000
from openai import OpenAI
client = OpenAI(base_url="http://localhost:30000/v1", api_key="none")
@sgl.function
test cases passed, vs 18 / 22 without
Published skills whose descriptions closely match this one, with their headline benchmark lift.
More skills from the same source and nearby in the registry, with their headline benchmark lift.