Changes / Feeds / nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8
nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8
No tracked key changed across the 2 recorded revisions since 2026-04-24. Head as last read: 0acd4b1c237eab45d3977b1b46631919e9a75774.
GPU memory to fine-tune it, card by card: Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8 on /m/.
A quantization of nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16: 6004 layers re-encoded; of the 1341 tensors left unquantized, 1318 byte-identical to the base, 23 cast to another dtype (BF16 to F32), so their hashes cannot compare values. Compared tensor by tensor from sha256 fingerprints: 0acd4b1c237e against nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16 at e5e9932441de. A hash is equal or not; it says nothing about what an edit does.
Pin it, and let a bot move the pin
from_pretrained("nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8", revision="0acd4b1c237eab45d3977b1b46631919e9a75774")
With the Renovate preset, a pull request moves this pin when the head moves, and links back here.
[](https://www.neurarch.com/changes-feeds/models/nvidia/Nemotron-3-Nano-Omni-30B-A3B-Reasoning-FP8.html)