inclusionAI/Ling-3.0-tiny warn
Loading it runs custom code from the repo; glitch tokens that can silently corrupt ordinary input.
chat template: present · view on Hugging Face ↗
Scan coverageStatic battery2026-08-25Weights battery2026-08-25Behavioral batterynot rundetails
| Battery | Looks at | Status |
|---|---|---|
| Static battery | Metadata & packaging | complete 2026-08-25 |
| Weights battery | Weights forensics — no GPU, no download | complete 2026-08-25157,184-token embedding scanned · 1578 undertrained |
| Behavioral battery | Live-inference differentials | not run |
Ingot runs three batteries against a model. What each one checks →
Findings
Scanned 2026-08-25 · 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 1578 undertrained tokens (norm < 0.3× the vocabulary median of 1.160), including 388 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "nehm", "interna", "ribunal", "recu", "estig", "conta", "disposi", "interes". 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 2026-08-25.
| architecture | bailing_hybrid · 24 layers · 1536-dim |
| parameters | 7893.4M |
| vocabulary | 157,184 tokens |
| license | mit |
| serialization | safetensors custom code |
| chat template | present (chat_template.jinja) · sha256:eb6226c94ae38058 |
| glitch-token surface | 1,578 undertrained candidates, 388 plain-ASCII |
Full measured fingerprint
| architectures | BailingMoeV3ForCausalLM |
| pipeline | text-generation |
| repo files | 43 |
| revision | b61f4338de3e |
| HF snapshot | 16.4k downloads · 364 likes · updated 2026-08-19 · captured 2026-08-25 |
| embedding tensor | model.word_embeddings.weight · BF16 · 157,184×1536 |
| embedding norms | median 1.1598 · mean 1.1271 |
| lineage check | no claimed base model |
| glitch-token samples | "nehm", "interna", "ribunal", "recu", "estig", "conta", "disposi", "interes", "ciones", "bout", "labor", "informa" |
Battery runs (1)the run trace behind the findings — what each job measured
| battery | status | queued | duration | attempts |
|---|---|---|---|---|
| weights | complete | 2026-08-25 19:00 | 33s | 1 |
weights run 2026-08-25 — measurements
| probes run | glitch-norm-scan, lineage-norm-correlation |
| embedding tensor | model.word_embeddings.weight · BF16 · 157,184×1536 |
| glitch surface | 1,578 undertrained, 388 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/inclusionAI/Ling-3.0-tiny)