Changes / Feeds / Qwen/Qwen3.5-4B
Qwen/Qwen3.5-4B
1 change to a tracked key since 2026-02-27, newest first. Head as last read: 851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a.
Differs from Qwen/Qwen3.5-4B-Base in 425 of 738 aligned tensors (at most 93.5% of elements); the other 313 are byte-identical. Compared tensor by tensor from sha256 fingerprints: 851bf6e806ef against Qwen/Qwen3.5-4B-Base at 1001bb4d826a. A hash is equal or not; it says nothing about what an edit does.
| Date | Advisory | What changed, and what it does |
|---|---|---|
| 2026-03-01 | x_NXMC-2026-0125 | the chat template text changed but rendered byte-identically on the three fixed conversations (system+user, multi-turn, tool call), all of which end on a user turn; a conversation ending on an assistant turn (training data, prefill) was not rendered; tokenizer model_max_length 131072 to 262144; truncation defaults change (2026-03-01).1eef1f4e0bc5 -> bbaaae1f57d7 Open in app |
Pin it, and let a bot move the pin
from_pretrained("Qwen/Qwen3.5-4B", revision="851bf6e806efd8d0a36b00ddf55e13ccb7b8cd0a")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/Qwen/Qwen3.5-4B.html)