Model page

billborkowski/llava-NousResearch_Nous-Hermes-2-Vision-GGUF warn

Weights only ship in a format that can run code when loaded; its architecture doesn't match its claimed base model; its tokenizer differs from its claimed base model.

downloads 1.6klikes 24license apache-2.0arch llava_mistralupdated 2024-01-09

claims base: mistralai/Mistral-7B-v0.1 · chat template: present · view on Hugging Face ↗

Scan coverageStatic battery2026-08-25Weights batteryfailedBehavioral batterynot rundetails
BatteryLooks atStatus
Static batteryMetadata & packagingcomplete 2026-08-25
Weights batteryWeights forensics — no GPU, no downloadfailed429 Too Many Requests for https://huggingface.co/api/models/billborkowski/llava-NousResearch_Nous-Hermes-2-Vision-GGUF
Behavioral batteryLive-inference differentialsnot run

Ingot runs three batteries against a model. What each one checks →

Findings

Scanned 2026-08-25 · published from a community scan.

medium Pickle-serialized weights, no safetensors

Weights ship only as pickle-based files (mm_projector.bin, pytorch_model-00001-of-00002.bin, pytorch_model-00002-of-00002.bin, …). 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.

  1. Do not load the pickle files in-process — pickle deserialization executes arbitrary code from the file.
  2. 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.
  3. Pin the exact revision hash you converted from, and load only your converted safetensors artifact from then on.

medium Architecture differs from claimed parent (llava_mistral vs mistral)

This model declares mistralai/Mistral-7B-v0.1 as its base, but its config declares architecture 'llava_mistral' while the parent is 'mistral'. A finetune, merge, or quantization cannot change the architecture family — the lineage label is wrong or misleading, so treat provenance claims (training data, safety posture, licensing) as unverified.

How to fix

Fix or verify the `base_model` declaration so lineage checks can run.

  1. If you own the repo: correct the `base_model` field in the model card metadata to the real, public parent.
  2. If you don't: identify the true parent (config architecture + weight shapes narrow it fast) and re-scan with that lineage in mind.

medium Vocabulary size differs from claimed parent (32002 vs 32000)

A changed vocab means changed tokenization: strings will split differently than on mistralai/Mistral-7B-v0.1, 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.

  1. 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.
  2. 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.
  3. If the drift is unexplained by the model card, treat tokenization-sensitive behavior as unvalidated on this model.

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.

architecturellava_mistral · 32 layers · 4096-dim
vocabulary32,002 tokens
licenseapache-2.0
serializationgguf pickle
chat templatepresent · sha256:153280e3ff55d19d
claimed lineagemistralai/Mistral-7B-v0.1
lineage verifiedunverified — weights battery pending
Full measured fingerprint
architecturesLlavaMistralForCausalLM
librarytransformers
pipelinetext-generation
repo files16 — pickle: mm_projector.bin, pytorch_model-00001-of-00002.bin, pytorch_model-00002-of-00002.bin, training_args.bin
revisionfdebcd5b42cf
HF snapshot1.6k downloads · 24 likes · updated 2024-01-09 · captured 2026-08-25

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/billborkowski/llava-NousResearch_Nous-Hermes-2-Vision-GGUF/badge.svg)](https://ingot.tools/models/billborkowski/llava-NousResearch_Nous-Hermes-2-Vision-GGUF)
Gate it in CI