Skip to content

Installing sglang[openai] does not have numpy requirement #70

Closed

Description

@darinkishore

I tried installing sglang[openai], ran the code in the readme:

import sglang as sgl
import snoop

@sgl.function
def tool_use(s, question):
    s += "To answer this question: " + question + ", "
    s += "I need to use a " + sgl.gen("tool", choices=["calculator", "web browser"]) + ". "
    if s["tool"] == "calculator":
        s += "The math expression is" + sgl.gen("expression")
    elif s["tool"] == "web browser":
        s += "The website url is" + sgl.gen("url")


@sgl.function
def tip_suggestion(s):
    s += (
        "Here are two tips for staying healthy: "
        "1. Balanced Diet. 2. Regular Exercise.\n\n"
    )

    forks = s.fork(2)
    for i, f in enumerate(forks):
        f += f"Now, expand tip {i+1} into a paragraph:\n"
        f += sgl.gen(f"detailed_tip", max_tokens=256, stop="\n\n")

    s += "Tip 1:" + forks[0]["detailed_tip"] + "\n"
    s += "Tip 2:" + forks[1]["detailed_tip"] + "\n"
    s += "In summary" + sgl.gen("summary")


@sgl.function
def text_qa(s, question):
    s += "Q: " + question + "\n"
    s += "A:" + sgl.gen("answer", stop="\n")


def driver_tool_use():
    state = tool_use.run(question="What is the capital of the United States?")
    print(state.text())
    print("\n")


def driver_tip_suggestion():
    state = tip_suggestion.run()
    print(state.text())
    print("\n")


def driver_batching():
    states = text_qa.run_batch(
        [
            {"question": "What is the capital of the United Kingdom?"},
            {"question": "What is the capital of France?"},
            {"question": "What is the capital of Japan?"},
        ],
        progress_bar=True
    )

    for s in states:
        print(s.text())
    print("\n")


def driver_stream():
    state = text_qa.run(
        question="What is the capital of France?",
        temperature=0.1,
        stream=True
    )

    for out in state.text_iter():
        print(out, end="", flush=True)
    print("\n")


if __name__ == "__main__":
    sgl.set_default_backend(sgl.OpenAI("gpt-3.5-turbo-instruct"))

    driver_tool_use()
    driver_tip_suggestion()
    driver_batching()
    driver_stream()

Fix: Add numpy + possibly more dependencies to the SGlang dependencies.

Activity

  1. merrymercy commented on Jan 21, 2024

    @merrymercy
    Collaborator

    Thanks for reporting this. Fixed by #71

  2. added a commit that references this issue on May 16, 2025
  3. added a commit that references this issue on Jun 5, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions