Top all-time · #21752
Osmosis
Company blog · osmosis.ai/blogaiprogramming
Osmosis is the company blog at osmosis.ai, writing about ai and programming. Since 2025, 3 different Hacker News readers have linked it 9 times, most often “Training over a thousand LoRA adapters at once”. No feed was found, so read it on the site itself.
Too few sampled comments (1) to say how often HN recommends it.?
Why HN links it
The linked material is a product announcement for an open source small language model, osmosis-mcp-4b, that connects any MCP client to any MCP server. No comment discusses the writing.
Summarised by Claude from a sample of 1 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Osmosis is a forward-deployed reinforcement learning platform that helps companies train task-specific AI models that outperform foundation models at a fraction of the cost. Build, deploy, and continuously improve intelligent systems with hands-on integration and real-time feedback.
Most-linked posts?
- Training over a thousand LoRA adapters at once2 people · 2 mentions · first linked 2026-06-02
What Hacker News says?
Hey HN! We’re Kasey and Andy from Osmosis ( https://osmosis.ai/ ). We’ve been playing with MCP recently and wanted to share a lightweight open source model that can connect any MCP client* to any MCP server! Check it out here: https://huggingface.co/osmosis-ai/osmosis-mcp-4b Right now, the only models that consistently work well for MCP are large and closed-source (e.g. 3.7 Sonnet, Gemini 2.5). Other models struggle…
Comment linking /
on “Open Source SLM Trained for MCP”Submitted Training over a thousand LoRA adapters at once as a story · 8 points
HN discussionSubmitted Open Source SLM Trained for MCP as a story · 8 points
HN discussionSubmitted LoRA's Limitations: Head-to-Head with Full RL as a story · 7 points
HN discussionSubmitted Show HN: LLM structured outputs hurt performance, we built a model to fix it as a story · 5 points
HN discussionSubmitted Unlocking LoRA Moe RL for Qwen3.5 as a story · 3 points
HN discussionSubmitted Code merging SLM exceeding performance of large models as a story · 3 points
HN discussionSubmitted Exploring Foundation Models' Tool-Use Efficacy as a story · 2 points
HN discussionSubmitted Training Thousands of LoRA Adapters at Once as a story · 1 points
HN discussion