HuggingFaceTB/SmolVLM-256M-Instruct warn
The chat template differs from its base model, which changes behavior; its tokenizer differs from its claimed base model.
claims base: HuggingFaceTB/SmolLM2-135M-Instruct, google/siglip-base-patch16-512 · chat template: present · view on Hugging Face ↗
Scan coverageStatic battery2026-08-22Weights battery2026-08-22Behavioral batterynot rundetails
| Battery | Looks at | Status |
|---|---|---|
| Static battery | Metadata & packaging | complete 2026-08-22 |
| Weights battery | Weights forensics: no GPU, no download | complete 2026-08-2249,280-token embedding scanned · 0 undertrained · lineage consistent |
| Behavioral battery | Live-inference differentials | not run |
Ingot runs three batteries against a model. What each one checks →
Findings
Scanned 2026-08-22 · published from a community scan.
medium Chat template differs from claimed parent
The chat template does not match HuggingFaceTB/SmolLM2-135M-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.
- Run `ingot patch <owner/model>` — the patch manifest carries the parent's template and applies it to a local copy's `tokenizer_config.json`.
- 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.
- 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 Vocabulary size differs from claimed parent (49280 vs 49152)
A changed vocab means changed tokenization: strings will split differently than on HuggingFaceTB/SmolLM2-135M-Instruct, which can shift behavior on identifiers, codes, and non-English text.
How to fixweight-level
Not patchable: the vocab size mirrors the embedding matrix in the weights. Verify the change was intentional.
- Do not edit `vocab_size` in config.json to "match the parent" — it must equal the embedding table in the shipped weights or the model won't load.
- Diff the tokenizers (`tokenizer.json` / added_tokens) against the parent to see what was added or removed, and test your own identifiers, codes, and non-English text through both.
- If the drift is unexplained by the model card, treat tokenization-sensitive behavior as unvalidated on this model.
info Embedding-norm glitch scan clean
No undertrained tokens found: every non-special token's embedding norm is above 0.3× the vocabulary median (3.058). The glitch-token data-corruption class has no candidate surface in this model.
info Weights consistent with claimed parent HuggingFaceTB/SmolLM2-135M-Instruct
Mean cosine similarity of 64 sampled token-embedding rows against HuggingFaceTB/SmolLM2-135M-Instruct is 0.997 — the weights plausibly descend from the declared base (relation: unspecified).
Check every checkpoint before it ships
Use the web app, API, CLI, or CI gate to scan candidate checkpoints and catch model drift before deployment. Public-model scans publish to the open database; paid plans add the volume needed for continuous checks.
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 HuggingFaceTB/SmolVLM-256M-Instruct
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 | idefics3 |
| parameters | 256.5M |
| vocabulary | 49,280 tokens |
| license | apache-2.0 |
| serialization | safetensors |
| chat template | present · sha256:830005085ddd94ce |
| claimed lineage | HuggingFaceTB/SmolLM2-135M-Instruct, google/siglip-base-patch16-512 |
| lineage verified | consistent vs HuggingFaceTB/SmolLM2-135M-Instruct — embedding-row cosine 0.997 |
| glitch-token surface | clean no undertrained tokens |
Full measured fingerprint
| architectures | Idefics3ForConditionalGeneration |
| library | transformers |
| pipeline | image-text-to-text |
| repo files | 38 |
| revision | 7e3e67edbbed |
| HF snapshot | 871.6k downloads · 398 likes · updated 2025-04-08 · captured 2026-08-21 |
| embedding tensor | model.text_model.embed_tokens.weight · BF16 · 49,280×576 |
| embedding norms | median 3.058 · mean 3.1033 |
| lineage check | consistent — cosine 0.9971 over 64 sampled rows vs HuggingFaceTB/SmolLM2-135M-Instruct |
Battery runs (1)the run trace behind the findings — what each job measured
| battery | status | queued | duration | attempts |
|---|---|---|---|---|
| weights | complete | 2026-08-21 07:43 | 20s | 1 |
weights run 2026-08-21 — measurements
| probes run | glitch-norm-scan, lineage-norm-correlation |
| embedding tensor | model.text_model.embed_tokens.weight · BF16 · 49,280×576 |
| glitch surface | 0 undertrained, 0 plain-ASCII |
| lineage check | consistent — cosine 0.9971 over 64 rows vs HuggingFaceTB/SmolLM2-135M-Instruct |
Verdict badge
Ship the verdict in your README — it always shows the latest published analysis:
[](https://ingot.tools/models/HuggingFaceTB/SmolVLM-256M-Instruct)