Changes / Feeds / Qwen/Qwen3.5-35B-A3B
Qwen/Qwen3.5-35B-A3B
4 changes to a tracked key since 2026-02-24, newest first. Head as last read: 59d61f3ce65a6d9863b86d2e96597125219dc754.
Differs from Qwen/Qwen3.5-35B-A3B-Base in 1458 of 1811 aligned tensors (at most 98.9% of elements); the other 353 are byte-identical. Compared tensor by tensor from sha256 fingerprints: 59d61f3ce65a against Qwen/Qwen3.5-35B-A3B-Base at 0f0813072d23. A hash is equal or not; it says nothing about what an edit does.
| Date | Advisory | What changed, and what it does |
|---|---|---|
| 2026-02-24 | x_NXMC-2026-0152 breaking | tokenizer_class "Qwen3_5Tokenizer" to "Qwen2Tokenizer" (2026-02-24).fd7fff71b103 -> aa891b984a2e Open in app |
| 2026-02-24 | x_NXMC-2026-0151 breaking | tokenizer_class "Qwen2Tokenizer" to "Qwen3_5Tokenizer" (2026-02-24).e2bbecc9fd0f -> fd7fff71b103 Open in app |
| 2026-02-24 | x_NXMC-2026-0150 | tokenizer model_max_length 131072 to 262144; truncation defaults change (2026-02-24).ebe7c66d760c -> e2bbecc9fd0f Open in app |
| 2026-02-24 | x_NXMC-2026-0149 | 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 (2026-02-24).41adbc1e5034 -> ce53c848c56d Open in app |
Pin it, and let a bot move the pin
from_pretrained("Qwen/Qwen3.5-35B-A3B", revision="59d61f3ce65a6d9863b86d2e96597125219dc754")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/Qwen3.5-35B-A3B.html)