DJKG/nano-kyrgyz warn
Weights only ship in a format that can run code when loaded; the weight files reference unusual code — review before loading.
Could not load this model from the Hugging Face API (private, gated, or nonexistent). Findings below are from our archive.
Scan coverageStatic battery2026-08-26Weights battery2026-08-26Behavioral batterynot rundetails
| Battery | Looks at | Status |
|---|---|---|
| Static battery | Metadata & packaging | complete 2026-08-26 |
| Weights battery | Weights forensics — no GPU, no download | complete 2026-08-26pickle audit: 14 non-standard global(s) |
| Behavioral battery | Live-inference differentials | not run |
Ingot runs three batteries against a model. What each one checks →
Findings
Scanned 2026-08-26 · published from a community scan.
medium Pickle-serialized weights, no safetensors
Weights ship only as pickle-based files (vision/chat_ckpt_v3.pt, vision/glyph_ckpt.pt, vision/vlm_ckpt.pt, …). Loading pickle executes arbitrary code from the file — prefer a safetensors release or load in a sandbox.
How to fix
Convert the weights to safetensors before loading them anywhere that matters.
- Do not load the pickle files in-process — pickle deserialization executes arbitrary code from the file.
- Convert locally in a sandbox: `pip install safetensors` and use `safetensors.torch.save_file` on a state dict loaded with `torch.load(..., weights_only=True)` (refuses most code-execution payloads), or use Hugging Face's `convert.py` space/script.
- Pin the exact revision hash you converted from, and load only your converted safetensors artifact from then on.
medium Pickle references non-standard globals
The pickle imports globals outside the standard torch/numpy/collections set: ultralytics.nn.tasks.DetectionModel, __builtin__.set, ultralytics.nn.modules.conv.Conv, ultralytics.nn.modules.block.C3k2, ultralytics.nn.modules.block.Bottleneck, ultralytics.nn.modules.block.C3k, +8 more. Common in full-model (non-state-dict) saves — each is code that runs at load time. Review before loading, or demand a safetensors release.
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 2026-08-26.
| license | apache-2.0 |
| serialization | gguf pickle |
| chat template | none |
Full measured fingerprint
| library | gguf |
| pipeline | text-generation |
| repo files | 25 — pickle: vision/chat_ckpt_v3.pt, vision/glyph_ckpt.pt, vision/vlm_ckpt.pt, vision/yolo_layout_best.pt |
| revision | 1419df5ce196 |
| HF snapshot | 152 downloads · 0 likes · updated 2026-08-25 · captured 2026-08-25 |
| pickle audit | vision/chat_ckpt_v3.pt, vision/glyph_ckpt.pt, vision/vlm_ckpt.pt, vision/yolo_layout_best.pt — 15 standard global(s), suspicious ultralytics.nn.tasks.DetectionModel, __builtin__.set, ultralytics.nn.modules.conv.Conv, ultralytics.nn.modules.block.C3k2, ultralytics.nn.modules.block.Bottleneck, ultralytics.nn.modules.block.C3k |
Battery runs (1)the run trace behind the findings — what each job measured
| battery | status | queued | duration | attempts |
|---|---|---|---|---|
| weights | complete | 2026-08-25 21:51 | 13s | 1 |
weights run 2026-08-25 — measurements
| probes run | gguf-metadata |
Verdict badge
Ship the verdict in your README — it always shows the latest published analysis:
[](https://ingot.tools/models/DJKG/nano-kyrgyz)