Try Ornith 1.0 Free: Chat in Your Browser

Experience Ornith 1.0 without installing anything. The Ornith 1.0-9B model runs on Hugging Face Spaces — just type a coding question and watch Ornith 1.0 reason through the solution.

Ornith 1.0-9B — Official Demo

The official Ornith 1.0-9B chat demo hosted by DeepReinforce AI on Hugging Face Spaces. Chat with the self-scaffolding coding model and see its reasoning trace in real time.

Open on HF ↗

Ornith 1.0 Demo Tips

What to Ask Ornith 1.0

  • • "Write a thread-safe LRU cache in Python"
  • • "Find the bug in this code: [paste code]"
  • • "Refactor this function to use async/await"
  • • "Explain how this regex works: [pattern]"

What to Expect from Ornith 1.0

  • • Ornith 1.0 shows a thinking trace before answering
  • • Responses may take 10-30 seconds on free tier
  • • Best results with specific coding questions
  • • This demo uses Ornith 1.0-9B (smallest variant)

Want More from Ornith 1.0?

The browser demo runs Ornith 1.0-9B on shared hardware. For faster responses and access to larger Ornith 1.0 models (35B MoE and 397B), run Ornith 1.0 locally on your own GPU.

Run Locally

Set up Ornith 1.0 on your GPU with vLLM, Ollama, or LM Studio for faster, private inference.

Compare Models

See which Ornith 1.0 model size fits your hardware — from 6GB VRAM to 8x GPU clusters.

Benchmarks

Full Ornith 1.0 benchmark results on SWE-Bench, Terminal-Bench, and ClawEval.