# sgl-project/sglang

> SGLang is an open-source inference framework for fast serving of large language and multimodal models, optimized for agentic workloads, RL rollouts, and large-scale deployment.

- Magnitude: 2.5 out of 10 — Steady
- Stars: 36,690 total · +14 stars today, ≈ 30 by evening
- Star trust: unusual star pattern (heuristic)
- Category: Language models · Language: Python · License: Apache-2.0 · Created: 2024-01-08 · Last push: 2026-10-01
- GitHub: https://github.com/sgl-project/sglang · Homepage: https://sglang.io · Page: https://gitnova.dev/en/r/sgl-project/sglang

## Useful for

- Launch a local inference server for Llama or DeepSeek using the lmsysorg/sglang Docker image
- Deploy a multimodal VLM model across multiple GPUs with load balancing
- Use SGLang as a rollout generation backend for RL training via Miles or verl

## Why it’s here

- 14 stars so far today, about 30 expected by the end of the day.
- GitHub Trending Python today: #19, +39 stars.
- About 18 forks a day — people are taking the code.
- Recent forks include notable developers: @bvolpato (1,763 followers), @geelen (1,536 followers).

## Star trust

Unusual star pattern (heuristic).
- Stars arrived in one batch on a single day (1,248) and stopped right after: 93 over the next two days. Real interest usually fades gradually.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 9,250
- Issues and pull requests: 41,557
- Watchers: 184
- Average over the last week: 45 per day
- Usual pace: 82 per day
- Stars in the last hour (measured): -1
- Latest release: v0.5.20 (2026-09-18)

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-09-02 … 2026-10-01: 515, 588, 1248, 39, 54, 60, 68, 66, 64, 43, 45, 53, 49, 59, 67, 44, 47, 34, 49, 70, 52, 42, 33, 25, 27, 41, 61, 78, 56, 14

## Spotted in now

- GitHub Trending Python today: #19, +39 stars

## Similar by description

1. **vllm-project/vllm-ascend** — 1.2 · Steady · Language models · Python · +2 stars today, ≈ 3 by evening
   A hardware plugin for vLLM that runs LLM inference on Ascend NPUs (Atlas A2/A3). It enables deployment of Transformer, MoE, embedding and multimodal models on Huawei Ascend hardware.
   Full card: https://gitnova.dev/en/r/vllm-project/vllm-ascend.md
2. **kserve/kserve** — 0.9 · Steady · DevOps & infrastructure · Go · +2 stars today, ≈ 3 by evening
   KServe is a platform for deploying and serving generative and predictive AI models on Kubernetes, supporting multiple frameworks, autoscaling, and a standardized inference protocol.
   Full card: https://gitnova.dev/en/r/kserve/kserve.md
3. **microsoft/onnxruntime** — 1.7 · Steady · Language models · C++ · +3 stars today, ≈ 7 by evening
   Cross-platform accelerator for ML inference and training. Runs models from PyTorch, TensorFlow, scikit-learn and others via the ONNX format with graph optimizations and hardware acceleration.
   Full card: https://gitnova.dev/en/r/microsoft/onnxruntime.md
4. **amitshekhariitbhu/ai-system-design** — 1.5 · Cooling · Learning & lists · Markdown · +10 stars today, ≈ 16 by evening
   A study guide to designing AI systems built on LLMs, RAG, and agents: from inference and GPUs to MCP, multi-agent systems, and interview preparation.
   Full card: https://gitnova.dev/en/r/amitshekhariitbhu/ai-system-design.md
5. **ml-explore/mlx** — 1.7 · Steady · Language models · C++ · +4 stars today, ≈ 7 by evening
   MLX is an array and machine learning framework from Apple, optimized for Apple silicon with unified memory and lazy computation. Its Python, C++, C, and Swift APIs mirror NumPy and PyTorch, simplifying model training and deployment.
   Full card: https://gitnova.dev/en/r/ml-explore/mlx.md

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-10-01 13:52 UTC, updated every 30 minutes.