83_Self_Improving_Loop

Hermes-Bonsai Karpathy Self-Improving Agent Loop

Stage 2 checkpoint for the Hermes/Bonsai Karpathy auto-research loop.

Last updated: 2026-04-05

This release is inspired by Andrej Karpathy's framing of self-improving training loops and auto-research. It contains the model artifact that worked, plus a concise model card explaining how it was produced and how to run it.

Overview

• Base model: Qwen3-8B-Base
• Training method: supervised fine-tuning via the Hermes/Karpathy loop
• Stage: Stage 2 — the checkpoint that worked
• Known limitation: Stage 3 exposed a learned-helplessness pattern on some tasks; that behavior is documented in the GitHub methodology repo
• License: Apache-2.0 for this release; the underlying base model license also applies to the inherited Qwen3-8B-Base components

What went into this checkpoint

• The loop-produced training curriculum and trace distillation pipeline
• 140 verified raw passes used as positive reinforcement for curriculum rebalancing and trace selection
  • These are Bonsai's own unedited outputs that passed teacher evaluation
• 10 domains covered across the build
• Validation signal from a mixed-domain batch

Domains covered

• memory_integration
• refusal_redirect
• self_correction
• agent_routing
• devops
• logic_puzzle
• code_debugging
• math
• architecture
• research_synthesis

Strongest domains

Best performance concentrated in:

• memory_integration
• refusal_redirect
• self_correction

Validation metrics

• Mixed-domain batch: 13/50 raw passes
• Raw pass rate: 26%
• This checkpoint is the stage 2 model that produced those verified passes

What's novel

Trained via a graduation protocol with teacher-guided validation, raw-pass reinforcement, and frontier failure analysis. The interesting contribution is the loop methodology; see GitHub for the full curriculum and training workflow.

GitHub methodology

The training loop, curriculum design, graduation protocol, and detailed methodology live here:

https://github.com/aurous37-lang/Hermes-Bonsai-Self-Improving-Agent-Loop

Files in this Hugging Face repo

• bonsai-8b-stage2-post-curriculum-q8.gguf — the shipped stage 2 checkpoint
• README.md — this model card
• LICENSE — Apache-2.0 license

How to use

Recommended working config from the stable local run:

• --ctx-size 40960
• --n-gpu-layers 37

llama.cpp

./llama-cli -m bonsai-8b-stage2-post-curriculum-q8.gguf \
  --ctx-size 40960 \
  -p "Explain the CAP theorem for a backend engineer."

llama-server

./llama-server -m bonsai-8b-stage2-post-curriculum-q8.gguf \
  --ctx-size 40960 \
  --n-gpu-layers 37 \
  --host 0.0.0.0 --port 8080

Then point your client at the local OpenAI-compatible endpoint exposed by llama-server.

Notes

• This is a release checkpoint, not the full training corpus.
• The GitHub repo contains the code and documentation needed to reproduce the loop.
• The Hugging Face repo contains the model artifact that ships from that loop.