amadan
paulca / paulca.com Public
| 1 | --- |
| 2 | model: GPT-OSS (RTX 3090) |
| 3 | wh: 2 |
| 4 | co2_g: 0.8 |
| 5 | comparison: A half-hour of a 5 W LED bulb |
| 6 | prompt: >- |
| 7 | “Switching from Claude to GPT-OSS on my gaming rig in the home office.” |
| 8 | --- |
| 9 | Switched to a local open‑source LLM running on an RTX 3090 GPU. No data |
| 10 | leaves home and energy per prompt is roughly 2 Wh. |
| 11 |