The week in brief
Behind it, Unsloth (+609) and Transformers (+580) followed.
One new tool joined the directory this week: Bonsai (Prism ML).
On Hacker News, “Beam: Reflection's 501B open-weight model” drew the most attention in this space (554 points, 174 comments). Among listed tools, Mistral AI's site was linked from 14 Hacker News stories over the past 90 days, more than any other tool here.
In research, “DiffusionART: Efficient, Stable and Scalable Reinforcement of Diffusion Models” collected the most upvotes on Hugging Face Daily Papers this week (147). Among new GitHub repositories created in the past 28 days, amitshekhariitbhu/ai-engineering-course leads with 526 stars.
Biggest GitHub star gainers
GitHub star counts for the 45 open-source tools we track, compared between Oct 5, 2026 and Oct 12, 2026 (155 hours apart).
| Tool | Stars before | Stars after | Gain | Change |
|---|---|---|---|---|
| PyTorch | 103,779 | 104,714 | +935 | 0.90% |
| Unsloth | 77,241 | 77,850 | +609 | 0.79% |
| Transformers | 166,982 | 167,562 | +580 | 0.35% |
| llama.cpp | 130,405 | 130,915 | +510 | 0.39% |
| Whisper | 110,008 | 110,370 | +362 | 0.33% |
| TensorFlow | 200,710 | 200,884 | +174 | 0.09% |
| nanoGPT | 63,569 | 63,732 | +163 | 0.26% |
| MLX | 28,661 | 28,733 | +72 | 0.25% |
New listings this week
| Tool | Category | Listed |
|---|---|---|
| Bonsai (Prism ML) | Open-Weight Models | Oct 08, 2026 |
Most discussed on Hacker News
The highest-scoring Hacker News stories in this space during the week.
| Story | Points | Comments | Discussion |
|---|---|---|---|
| Beam: Reflection's 501B open-weight model | 554 | 174 | thread |
| The gap in shared understanding of LLM capability is widening | 500 | 731 | thread |
| Dust: Pretraining Transformers Without Backpropagation | 279 | 82 | thread |
| Port of the TypeScript compiler, checker and lsp to Rust, by LLM | 117 | 222 | thread |
| LLMs may have helped my RSI | 102 | 59 | thread |
| Write Like It's 1866: LLMs Relearn Telegraphese | 96 | 58 | thread |
Listed tools with the most Hacker News links
Hacker News stories linking to each tool's own website over the past 90 days.
Notable new papers
The most-upvoted papers on Hugging Face Daily Papers and the newest relevant arXiv preprints from this week.
- DiffusionART: Efficient, Stable and Scalable Reinforcement of Diffusion Models
- S-Space: Exploring Spatial Workspace in Multimodal Language Models
- Reflex: Visually Grounded Reactive Humanoid Control
- SoftwareGym: Scaling Computer-Use Agent Training with Diverse Synthesized Software Environments
- EmbodiedCLBench: Evaluating Continual Learning for Self-Evolving Embodied Agents
- Task Context Shapes Component Roles in Language Models
- On the Design Space of Continuous Diffusion Language Models
- Can Language Models Learn When to Abstain? Policy Learning and Calibrated Answer Selection
- QTestBench: A Test Framework for Evaluating LLM-Generated Quantum Code
- Tracing and Taming Hallucinated Terms via Subsequence-Association Filtering and Breadth-Aware Selection for LLM Query Expansion
New GitHub repos in LLM research
Repositories created in the 28 days before Oct 12, 2026, ranked by stars.
| Repository | Stars | Language | Description |
|---|---|---|---|
| amitshekhariitbhu/ai-engineering-course | 526 | Markdown | AI Engineering Course - A free and complete AI Engineering Course to learn AI Engineering step by step - from Machine Learning, Neural Networks, and Transformers to LLMs, Fine-Tuning, RAG, AI Agents, |
| liyupi/ai-model-world | 226 | TypeScript | AI 大模型世界，把 556 个大模型拟人化成像素小人的可视化站点。进来就能看到此刻谁最聪明、谁最会写代码、谁最便宜、谁刚发布，往下是国内与国外分区的厂商广场、完整的发布时间线和多维排行榜。搜索认模型名、厂商和能力，输入「多模态」会直接列出全部多模态模型。数据取自 Epoch AI、models.dev、LiveBench 与 Hugging Face，每小时自动同步，所有文案由真实数据生成，不调 |
| Sidiora-Labs/centra-gideon-agent | 217 | Python | The companion AI agent that learns, adapts and gets the work done no matter the task |
| Dreamer-Toby/STEPQuant | 101 | Python | STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization |
| NiuTrans/AML-Notes | 50 | TeX | 《高级机器学习》：面向生成式人工智能时代的机器学习教材，涵盖强化学习、大模型训练与对齐、生成建模和多模态学习等内容。 |
How we compile this report
Every figure comes from a public source: star counts from the GitHub API, story scores and links from Hacker News (via Algolia), papers from arXiv and Hugging Face Daily Papers, and new repositories from GitHub search. Hacker News link counts include stories whose link points to the tool's own domain. Listing dates are the day a tool joined Axiomi. Numbers are shown as recorded; if a figure is unavailable, it is left out rather than estimated.