Mirror of Qwen/Qwen3-VL-2B-Instruct @ 89644892e4d8

This repository is a byte-exact mirror, kept by the AutoShade project so that a pinned model stays installable if its upstream host changes. It is not a new model, a fine-tune or a re-quantisation: every file here was copied from the upstream revision below and accepted only after its SHA-256 matched the checksum AutoShade's sidecars already enforce at download time. AutoShade fetches from here first and falls back to the upstream repository; either way the same checksum decides whether the bytes are used.

upstream Qwen/Qwen3-VL-2B-Instruct
upstream revision 89644892e4d85e24eaac8bacfd4f463576704203
licence apache-2.0, as declared by the upstream model card

Licence. The bytes are passed on under the licence upstream publishes them under, unchanged. That repository ships no licence file: like most model repositories it declares its licence through the card metadata, and the declaration above is the same one, read from the Hub API for that exact repo. Whatever the upstream licence permits or restricts applies to this copy too — mirroring changes where the bytes live, never the terms they carry.

Provenance. The file list here, with its sizes and LFS digests, is the evidence: each one equals the pin in AutoShade's sidecar for Qwen/Qwen3-VL-2B-Instruct at revision 89644892e4d85e24eaac8bacfd4f463576704203.

Downloads last month
57
Safetensors
Model size
2B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support