microsoft/Phi-3.5-vision-instruct warn
chat template: present · view on Hugging Face ↗
Scan coverage
Ingot runs three batteries against a model. What each one checks →
| Battery | Looks at | Status |
|---|---|---|
| Static battery | Metadata & packaging | complete 2026-08-22 |
| Weights battery | Weights forensics — no GPU, no download | complete 2026-08-2232,064-token embedding scanned · 536 undertrained |
| Behavioral battery | Live-inference differentials | not run |
Findings
Scanned 2026-08-22 · published from a community scan.
medium Repo ships executable Python (trust_remote_code)
The repository contains custom code files that run in-process when loaded with trust_remote_code=True. Pin the revision hash and review the code before loading.
How to fix
Review and pin the custom code; never float on `main` with trust_remote_code=True.
- Read every `.py` file in the repo before first load — this code runs in your process.
- Pin the revision: `from_pretrained(model_id, revision="<commit sha>", trust_remote_code=True)` so a later push can't swap the code under you.
- Prefer a version of the architecture already in `transformers` if one exists, which removes the remote-code requirement entirely.
medium Undertrained (glitch) token surface in vocabulary
Embedding-norm scan flagged 536 undertrained tokens (norm < 0.3× the vocabulary median of 2.153), including 96 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "<0xFC>", "<0xFB>", "<0xFF>", "<0xFD>", "<0xFA>", "<0xFE>", "Mediabestanden", "autorytatywna". 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.
- 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.
- 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.
- 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 profile of this model: measured weights-and-metadata facts, rebuilt on every scan and battery run. Updated 2026-08-22.
| architecture | phi3_v · 32 layers · 3072-dim |
| parameters | 4146.6M |
| vocabulary | 32,064 tokens |
| license | mit |
| serialization | safetensors custom code |
| chat template | present · sha256:79e4ca4b7d902dc7 |
| glitch-token surface | 536 undertrained candidates, 96 plain-ASCII |
Full measured fingerprint
| architectures | Phi3VForCausalLM |
| library | transformers |
| pipeline | image-text-to-text |
| repo files | 21 |
| revision | 12b77fb40b63 |
| HF snapshot | 991.5k downloads · 738 likes · updated 2025-12-10 · captured 2026-08-21 |
| embedding tensor | model.embed_tokens.weight · BF16 · 32,064×3072 |
| embedding norms | median 2.1534 · mean 2.0441 |
| lineage check | no claimed base model |
| glitch-token samples | "<0xFC>", "<0xFB>", "<0xFF>", "<0xFD>", "<0xFA>", "<0xFE>", "Mediabestanden", "autorytatywna", "Webachiv", "regnigaste", "tatywna", "Jegyzetek" |
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.
| battery | status | queued | duration | attempts |
|---|---|---|---|---|
| weights | complete | 2026-08-21 07:43 | 11s | 1 |
weights run 2026-08-21 — measurements
| probes run | glitch-norm-scan, lineage-norm-correlation |
| embedding tensor | model.embed_tokens.weight · BF16 · 32,064×3072 |
| glitch surface | 536 undertrained, 96 plain-ASCII |
| lineage check | not checked (no claimed base model) |
Verdict badge
Ship the verdict in your README — it always shows the latest published analysis:
[](https://ingot.tools/models/microsoft/Phi-3.5-vision-instruct)