Changes / Feeds / Qwen/Qwen3-4B-Base
Qwen/Qwen3-4B-Base
2 changes to a tracked key since 2025-04-28, newest first. Head as last read: 906bfd4b4dc7f14ee4320094d8b41684abff8539.
| Date | Advisory | What changed, and what it does |
|---|---|---|
| 2025-05-20 | x_NXMC-2025-0119 breaking | EOS changed from "<|im_end|>" to "<|endoftext|>"; anything fine-tuned or served against the old value stops at a different token (2025-05-20).08a151eb7f61 -> 08238297d7a5 |
| 2025-04-28 | x_NXMC-2025-0118 breaking | the chat template changed and renders differently (multi-turn, tool-call); prompts rendered with the old one no longer match what the model now expects, so cached prompts and eval sets need re-rendering (2025-04-28).2fcc0a0516f9 -> 08a151eb7f61 |
Pin it, and let a bot move the pin
from_pretrained("Qwen/Qwen3-4B-Base", revision="906bfd4b4dc7f14ee4320094d8b41684abff8539")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/Qwen3-4B-Base.html)