opus-4b-py-step145-2026-05-01

LoRA adapter (rank 32) trained with RL on a custom Opus-Magnum-style motion-planning task using the py answer representation. Snapshot at training step 145 / 300.

Source training run

wandb: 8y5y2yju
tinker checkpoint: tinker://6e8f08e2-b349-574e-917f-504f81718aa5:train:0/sampler_weights/000145
distances: 1, 2, 3
task types: move, transmute (no bond)
learning rate: 1e-5
group size: 8, groups per batch: 16
renderer: qwen3_5_disable_thinking

Usage

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

base = "Qwen/Qwen3.5-4B"
adapter = "maxbittker/opus-4b-py-step145-2026-05-01"
tok = AutoTokenizer.from_pretrained(base)
model = AutoModelForCausalLM.from_pretrained(base, device_map="auto")
model = PeftModel.from_pretrained(model, adapter)

Downloads last month: 13

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for maxbittker/opus-4b-py-step145-2026-05-01

Base model

Qwen/Qwen3.5-4B-Base

Finetuned

Qwen/Qwen3.5-4B

Adapter

(191)

this model