Changes / Feeds / ibm-granite/granite-vision-4.1-4b
ibm-granite/granite-vision-4.1-4b
3 changes to a tracked key since 2026-04-16, newest first. Head as last read: 37d591f06319e8f1638b5adcf58bdf50e0f84f7a.
GPU memory to fine-tune it, card by card: granite-vision-4.1-4b on /m/.
| Date | Advisory | What changed, and what it does |
|---|---|---|
| 2026-07-13 | x_NXMC-2026-0007 breaking | vocab_size changed from 100353 to 100352; the embedding and output head change shape, so adapters and merges made against the old vocabulary no longer load cleanly; derived parameters 3.32B to 3.32B (-2.6K, -0.00 GB of bf16 weights) (2026-07-13).6eada830c190 -> 37d591f06319 |
| 2026-07-12 | x_NXMC-2026-0006 breaking | tokenizer_class "GPT2Tokenizer" to "PreTrainedTokenizerFast" (2026-07-12).b593b0ae76f8 -> 371010331907 |
| 2026-04-28 | x_NXMC-2026-0005 | the chat template text changed but rendered byte-identically on the fixed conversation set (system+user, multi-turn, tool call) (2026-04-28).3556c946df5c -> b593b0ae76f8 |
Pin it, and let a bot move the pin
from_pretrained("ibm-granite/granite-vision-4.1-4b", revision="37d591f06319e8f1638b5adcf58bdf50e0f84f7a")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/ibm-granite/granite-vision-4.1-4b.html)