Model comparison
pick_cube_lora_1.2w vs o3
Capability radar, value bars, and side-by-side posture.
pick_cube_lora_1.2w
Not scored
value score
o3
93
value score
pick_cube_lora_1.2w ctx
—
o3 ctx
200K
Capability radar
Value · context · multimodal · openness · speed posture
Value head-to-head
- o393
Attribute tape
| Field | pick_cube_lora_1.2w | o3 |
|---|---|---|
| Developer | KaiyueChen | OpenAI |
| Context | — | 200K |
| Modalities | Text, Code | |
| Openness | — | Proprietary Model |
| Speed | — | — |
| Price | — | $10/1M in · $40/1M out |
| Value score | — | 93 |
| Summary | A task-specific LoRA adapter checkpoint published by KaiyueChen on Hugging Face, trained to specialize a base model on a narrow 'pick cube' robotic or vision task. It is a small set of low-rank weight updates intended to be loaded on top of a separate base model rather than run standalone. | OpenAI reasoning model focused on hard problems. |