Model page

ornith-ai/Ornith-1.0-9B warn

downloads 2.3Mlikes 536license mitarch qwen3_5params 1.5Mupdated 2026-06-25

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 2026-08-22
Weights batteryWeights forensics — no GPU, no downloadcomplete 2026-08-22248,320-token embedding scanned · 1582 undertrained
Behavioral batteryLive-inference differentialsnot run

Findings

Scanned 2026-08-22 · published from a community scan.

medium Undertrained (glitch) token surface in vocabulary

Embedding-norm scan flagged 1582 undertrained tokens (norm < 0.3× the vocabulary median of 0.885), including 396 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "tedothi", "ForCanBeConvertedToF", "ForCanBeConverted", "szexf", "Kinhted", "xfabl", "PostalCodesNL", "useRalative". 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.

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.

architectureqwen3_5 · 4096-dim
parameters1.5M
vocabulary248,320 tokens
licensemit
serializationsafetensors
chat templatenone
glitch-token surface1,582 undertrained candidates, 396 plain-ASCII
Full measured fingerprint
architecturesQwen3_5ForConditionalGeneration
librarytransformers
pipelinetext-generation
repo files18
revision83dc1f5e24ef
HF snapshot2.3M downloads · 535 likes · updated 2026-06-25 · captured 2026-08-21
embedding tensormodel.language_model.embed_tokens.weight · BF16 · 248,320×4096
embedding normsmedian 0.8851 · mean 0.8514
lineage checkno claimed base model
glitch-token samples"tedothi", "ForCanBeConvertedToF", "ForCanBeConverted", "szexf", "Kinhted", "xfabl", "PostalCodesNL", "useRalative", "tarsker", "tarskereso", "ejahter", "useRal"

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 07:4283s1
weights run 2026-08-21 measurements
probes runglitch-norm-scan, lineage-norm-correlation
embedding tensormodel.language_model.embed_tokens.weight · BF16 · 248,320×4096
glitch surface1,582 undertrained, 396 plain-ASCII
lineage checknot checked (no claimed base 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/ornith-ai/Ornith-1.0-9B/badge.svg)](https://ingot.tools/models/ornith-ai/Ornith-1.0-9B)
Gate it in CI