GitHub Projects Community (@GithubProjects) header
GitHub Projects Community (@GithubProjects) profile picture

We're sharing/showcasing best of @github projects/repos. Follow to stay in loop. Promoting Open-Source Contributions. UNOFFICIAL, but followed by github

San Francisco, CA
Joined October 2020
Photo by GitHub Projects Community (@GithubProjects)Photo by GitHub Projects Community (@GithubProjects)Photo by GitHub Projects Community (@GithubProjects)Photo by GitHub Projects Community (@GithubProjects)Photo by GitHub Projects Community (@GithubProjects)Photo by GitHub Projects Community (@GithubProjects)Photo by GitHub Projects Community (@GithubProjects)Photo by GitHub Projects Community (@GithubProjects)Photo by GitHub Projects Community (@GithubProjects)Photo by GitHub Projects Community (@GithubProjects)
Building an AI workflow platform or agent workspace? FlowGram hands you the canvas, node forms, and variable scoping already built, so you can spend your time on the parts that are yours. osp.fyi/flowgram-ai
2
4
2,428
One open-source downloader for HTTP, BitTorrent, YouTube, Bilibili, and more, on desktop and Android. Ghost Downloader 3 grabs videos and torrents from many sources in one app, no code needed.
3
4
3
37
5,206
4 projects readers kept coming back to this week. The top one swaps out a well-known product, and you can try it in a single step. ↓
1
1
17
8,474
3. airgorah — A WiFi security auditing software • Combines WiFi traffic capture, client discovery, deauthentication Why try it: It's open source (MIT), so you can try it today and see exactly how it works osp.fyi/airgorah
1
4
1,664
4. artex — Run an autonomous penetration test against in-scope targets, with findings tied to captured HTTP evidence • Open source and free to inspect Why try it: It's open source (AGPL-3.0), so you can try it today and see exactly how it works osp.fyi/artex
5
1,595
Unlimited’ AI tools usually still come with limits somewhere. Dreamina is taking a different approach. Dreamina’s REAL Unlimited offer for its top models is now live: up to 30 days of Seedance 2.5, plus up to 60 days of GPT Image 2.5. On a qualifying plan, eligible generations use exactly 0 credits. No rationing every attempt. No worrying about whether the next generation is worth the credits. #DreaminaPartner
3
3
38
11,485
The obvious question is: “Is Dreamina Unlimited actually unlimited?” The demo here shows Seedance 2.5 generations running back to back across different shots, angles, and ideas. The result: every eligible generation used exactly 0 credits. Generation stayed fast, and the output quality remained consistent throughout. No credit anxiety. Just more room to keep iterating.
1
2
1,451
The mindset before Unlimited: “I better save my credits just in case...” After Unlimited: “ONE MORE TAKE. AND ANOTHER. KEEP THEM COMING.” More experiments. Fresh ideas. Different angles. When eligible generations aren’t consuming credits, there’s a lot more freedom to keep refining instead of settling for the first draft. #DreaminaCanvas
1,278
Turn photos of real-world objects and environments into detailed 3D models. OpenMVS is an open-source 3D reconstruction library that transforms camera-aligned images into dense point clouds, refined meshes, and textured 3D models. Explore it here: osp.fyi/openmvs
3
6
1
120
9,599
Running LLMs on specialized AI chips shouldn't be a headache. TrnRT cuts through the complexity, making it easier to deploy and serve AI models through an OpenAI-compatible API.
3
3
40
7,108
Documents and watch your whole homelab - selfhostable Explore it here: osp.fyi/homelable
4
10
175
13,206
| ̄ ̄ ̄ ̄ ̄ ̄ ̄ ̄ ̄ ̄ ̄ ̄ ̄ ̄ ̄ ̄ ̄| | Don't Push To Production On Friday | |_________________| . 😎 |\👐 / \ ━━━━━┓ \\ ┓┓┓┓┓┃ ┓┓┓┓┓┃ ヽ😩ノ ┓┓┓┓┓┃   /  ┓┓┓┓┓┃ ノ)  ┓┓┓┓┓┃ ^^^^^^^^^^^Production^^^^
17
54
6
357
16,096
Run Windows apps and installers directly on Linux, without a virtual machine or complicated setup. Explore it here: osp.fyi/raven
3
3
1
86
10,133
SenseNova-Loop-9B explores a different way to scale an open-weight model: reuse computation instead of storing a much deeper stack of unique layers. Its shared 24-layer middle block runs twice, producing 56 layer computations while storing 32 distinct layers. The idea is to give a ~9B model more processing depth without expanding its parameter count in the usual way. It also supports a 512K multimodal context, coding, reasoning and multi-step tool use, with deployment options for Transformers, LightLLM, SGLang and vLLM. Explore the architecture and code: github.com/OpenSenseNova/Sen…
#SenseNova-Loop-9B is officially here — now #OpenSource! Just 9B parameters, competitive with Qwen3.8-27B across multiple benchmarks. 🔹 Big gains over Preview: NL2Repo 19.91 → 41.36; AutomationBench 14.33 → 36.67; GAIA 37.86 → 69.90 🔹 Loop + mHC: 32 unique layers, 56 layer executions 🔹 Day-0 4-bit AWQ weights (~4.9GB), with VRAM needs as low as ~9 GiB at 64K context 🔹 Up to 512K context, with text and image inputs See what a 9B model can build: animated posters, interactive websites and 3D scenes. 💻 GitHub: github.com/OpenSenseNova/Sen… 🤗 HF: huggingface.co/collections/s… 👾 Discord: discord.com/invite/BuTXPHmQu…
6
2
1
51
30,481
Manage AI-powered work requests with built-in review and approval workflows. Open-source, self-hosted, and designed to keep you in control. Explore it here: osp.fyi/soduwork
4
6
1
60
9,704