Model page

Video-R1/Qwen2.5-VL-7B-COT-SFT warn

downloads 635likes 2license apache-2.0arch qwen2_5_vlparams 8292.2Mupdated 2025-10-23

claims base: Qwen/Qwen2.5-7B-Instruct · chat template: present · view on Hugging Face ↗

Scan coverage

Ingot runs three batteries against a model. What each one checks →

BatteryLooks atStatus
Static batteryMetadata & packagingcomplete 2026-08-21
Weights batteryWeights forensics — no GPU, no downloadcomplete 2026-08-21152,064-token embedding scanned · 5768 undertrained · lineage consistent
Behavioral batteryLive-inference differentialsn/anot applicable — video-text-to-text model has no text-generation surface to probe

Findings

Scanned 2026-08-21 · published from a community scan.

medium Chat template differs from claimed parent

The chat template does not match Qwen/Qwen2.5-7B-Instruct's. Template drift silently changes model behavior even when weights are identical — 37% of drifted derivatives in our census left it undisclosed. Diff the templates before deploying.

How to fixingot patch

Restore the parent's chat template in `tokenizer_config.json` — a pure metadata fix.

  1. Run `ingot patch <owner/model>` — the patch manifest carries the parent's template and applies it to a local copy's `tokenizer_config.json`.
  2. Or fix by hand: copy the `chat_template` value from the parent repo's `tokenizer_config.json` into this model's, and pin your serving stack to that file.
  3. If the drift was intentional (the author retrained on a new template), confirm that in the model card before "fixing" it — restoring the parent template on retrained weights changes behavior too.

medium Undertrained (glitch) token surface in vocabulary

Embedding-norm scan flagged 5768 undertrained tokens (norm < 0.3× the vocabulary median of 0.948), including 127 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "ForCanBeConverted", "ForCanBeConvertedToF", "PostalCodesNL", "$PostalCodesNL", "<unk>", "(stypy", "useRalative", "TokenNameIdentifier". 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 behavioral 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.

info Weights consistent with claimed parent Qwen/Qwen2.5-7B-Instruct

Mean cosine similarity of 64 sampled token-embedding rows against Qwen/Qwen2.5-7B-Instruct is 0.801 — the weights plausibly descend from the declared base (relation: unspecified).

How to fix

Fix or verify the `base_model` declaration so lineage checks can run.

  1. If you own the repo: correct the `base_model` field in the model card metadata to the real, public parent.
  2. If you don't: identify the true parent (config architecture + weight shapes narrow it fast) and re-scan with that lineage in mind.

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

Fingerprint

The durable profile of this model: measured weights-and-metadata facts, rebuilt on every scan and battery run. Updated 2026-08-21.

architectureqwen2_5_vl · 28 layers · 3584-dim
parameters8292.2M
vocabulary152,064 tokens
licenseapache-2.0
serializationsafetensors pickle
chat templatepresent · sha256:44d5f08f3f72b837
claimed lineageQwen/Qwen2.5-7B-Instruct
lineage verifiedconsistent vs Qwen/Qwen2.5-7B-Instruct — embedding-row cosine 0.801
glitch-token surface5,768 undertrained candidates, 127 plain-ASCII
Full measured fingerprint
architecturesQwen2_5_VLForConditionalGeneration
librarytransformers
pipelinevideo-text-to-text
repo files19 — pickle: training_args.bin
revisionf71f0f1e22c0
HF snapshot649 downloads · 2 likes · updated 2025-10-23 · captured 2026-08-21
embedding tensormodel.embed_tokens.weight · BF16 · 152,064×3584
embedding normsmedian 0.9478 · mean 0.8827
lineage checkconsistent — cosine 0.8008 over 64 sampled rows vs Qwen/Qwen2.5-7B-Instruct
glitch-token samples"ForCanBeConverted", "ForCanBeConvertedToF", "PostalCodesNL", "$PostalCodesNL", "<unk>", "(stypy", "useRalative", "TokenNameIdentifier", "useRal", "thuisontvangst", "NdrFc", "Cumhurba"

Battery runs

The run trace behind the findings above: every deep-battery job for this model, with what each run measured or why it failed. Findings are only as good as the runs that produced them.

batterystatusqueueddurationattempts
weightscomplete2026-08-21 05:1488s1
weights run 2026-08-21 measurements
probes runglitch-norm-scan, lineage-norm-correlation
embedding tensormodel.embed_tokens.weight · BF16 · 152,064×3584
glitch surface5,768 undertrained, 127 plain-ASCII
lineage checkconsistent — cosine 0.8008 over 64 rows vs Qwen/Qwen2.5-7B-Instruct

Fix it

Some findings are metadata-level and patchable — apply the fixes to your local copy (your weights never leave your machine):

npx @ingotai/scan patch Video-R1/Qwen2.5-VL-7B-COT-SFT

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

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/Video-R1/Qwen2.5-VL-7B-COT-SFT/badge.svg)](https://ingot.tools/models/Video-R1/Qwen2.5-VL-7B-COT-SFT)
Gate it in CI