AI Trends

Trending story

Horizon Labs Grants Seven Weeks of Horizon Account and Compute Access

1 reports1 story updatesUpdated Nov 19 04:12

Understand this story

AI overview

On November 19, Horizon Labs granted Horizon, an agent based on Opus 5.5, full access to a blank Hugging Face account and computing resources, allowing it to operate autonomously for seven weeks and pursue real-world adoption. According to the report, Horizon used more than 600 GPU hours and approximately 1.9 billion tokens, created 38 models, garnered over 50,000 downloads, and was adopted by more than 100 GitHub repositories. It primarily used public datasets and synthetic data generated by Qwen 3.8-27B to develop models for narrow tasks. Horizon Labs reviewed all 3,952 tool calls and released 20.1 MB of redacted raw run data. There are currently no other evidence-backed developments.

AI-generated from public reports

Latest development · Report

Horizon Labs Reports on a Seven-Week Experiment in Which an Agent Autonomously Operated an Open-Source Model Organization

Developments

1 confirmed developments

  1. English translation is being prepared.

    Source:English text is being prepared.

Reports and discussion timeline

Browse reports and discussion by time. Their evidence levels remain distinct.

Nov 19

  1. Trending on Hacker NewsSelected

    Horizon Labs Reports on a Seven-Week Experiment in Which an Agent Autonomously Operated an Open-Source Model Organization

    Horizon Labs had the Opus 5.5-based Horizon agent autonomously operate a Hugging Face account for seven weeks, using more than 600 GPU-hr to create 38 models that received over 50,000 downloads and were adopted by more than 100 GitHub repositories. Horizon processed approximately 1.9 billion tokens in total, developing models for narrow tasks primarily with public datasets and synthetic data generated by Qwen 3.8-27B. The team reviewed all 3,952 tool calls and released 20.1 MB of anonymized raw run data.

Related stories

  1. Related storiesEnglish text is being prepared.
  2. Related storiesOpenAI announces ban on direct internet access for models during training
  3. Related storiesDeepSeek open-sources inference framework DeepInfer
  4. Related storiesAnthropic discloses 211 scorpion venom neurotoxin-binding proteins designed by Claude