Model page

Qwen/Qwen3-VL-8B-Instruct warn

downloads 5.3Mlikes 1.1klicense apache-2.0arch qwen3_vlparams 8767.1Mupdated 2025-10-15

chat template: present · view on Hugging Face ↗

Ingot findings

Static battery clean: safetensors weights, license declared, no template/tokenizer drift detected. Deep battery not yet run. Weights battery: embedding-norm scan over 151936 tokens (BF16, 4096-dim) found 2943 undertrained candidates, 100 plain-ASCII. Scanned 2026-08-20 (published from a community scan).

medium Undertrained (glitch) token surface in vocabulary

Embedding-norm scan flagged 2943 undertrained tokens (norm < 0.3× the vocabulary median of 1.396), including 100 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "$PostalCodesNL", "ForCanBeConvertedToF", "PostalCodesNL", "ForCanBeConverted", "useRalative", "useRal", "NdrFc", "sexkontakte". In models where this class was tested behaviorally, such tokens silently rewrote user input into confident, schema-valid, wrong output. These are candidates from the weights alone; behavioral confirmation requires the GPU deep battery.

How to fixruntime guardweight-level

Keep the affected token strings out of the model's input — the scan-derived runtime guard carries this model's exact blocklist.

  1. Fetch this model's guard artifact (`/api/v1/guard/<owner>/<model>`): the confirmed corrupting tokens and the low-norm candidate list, derived from the published scan.
  2. Screen inbound text with it (the `@ingotai/guard` package is a reference implementation) and route flagged records to a different model or human review — verbatim-copy tasks on flagged strings are the failure mode.
  3. The underlying cause is undertrained embeddings in the weights; a true fix is weight-level (continued pretraining on the affected tokens) — that is not a patch, it's a training job.

Remediation guidance addresses the documented findings only. It is evidence-driven repair, not a safety certification of the model.

Fingerprint

The durable weights-and-metadata profile of this model, rebuilt on every scan and deep-battery run. Updated 2026-08-20.

architectureqwen3_vl
parameters8767.1M
vocabulary151,936 tokens
licenseapache-2.0
serializationsafetensors
chat templatepresent · sha256:3636d0f0bd6bef02
glitch-token surface2,943 undertrained candidates, 100 plain-ASCII
Full fingerprint
architecturesQwen3VLForConditionalGeneration
librarytransformers
pipelineimage-text-to-text
repo files16
revision0c351dd01ed8
HF snapshot5.3M downloads · 1.1k likes · updated 2025-10-15 · captured 2026-08-20
embedding tensormodel.language_model.embed_tokens.weight · BF16 · 151,936×4096
embedding normsmedian 1.3956 · mean 1.3226
lineage checkno claimed base model
glitch-token samples"$PostalCodesNL", "ForCanBeConvertedToF", "PostalCodesNL", "ForCanBeConverted", "useRalative", "useRal", "NdrFc", "sexkontakte", "webElementX", "davidjl", "sextreffen", "-vesm"

Verdict badge

Ship the verdict in your README — it always shows the latest published analysis:

Ingot verdict: warn

[![Ingot scan](https://ingot.tools/api/v1/models/Qwen/Qwen3-VL-8B-Instruct/badge.svg)](https://ingot.tools/models/Qwen/Qwen3-VL-8B-Instruct)
Gate it in CI