run log

Gemma 4 12B Unified read pieces, but did not close the work

Gemma 4 12B Unified was added as a local LM Studio run on the Mac mini M4. It found parts of the paperwork, but the final result was harsh: 10% Practical Score, 0/5 resolved generated-image cases, workflow loops, and no parseable City Plan SVG.

Gemma 4 12B Unified via LM Studio benchmark note infographic
10% Practical score
0/5 Resolved
1/5 Core pass
SVG failed Visual sample

Gemma 4 12B Unified is now in the Local Model Bench data.

This was not a flattering run. The model did not look completely blind to the documents. It extracted and matched enough to pass some visible checks. But the benchmark is not asking for a plausible reading. It is asking for a finished artifact that survives the oracle.

That is where the run broke.

What ran

The model was run locally through LM Studio as `google/gemma-4-12b` on a Mac mini M4 with 64 GB unified memory.

It ran the five generated-image Paperwork Trial cases, the four agentic Paperwork Workflow cases, and the City Plan SVG sanity check.

The generated-image cases use synthetic invoice scans plus bank exports, vendor records, purchase orders, and exact `audit_result.json` checks. The workflow cases add messy folders, protected source files, intermediate artifacts, and proof files.

The headline result

On the generated-image Paperwork Trial, Gemma 4 12B Unified scored 10% Practical Score: zero of five cases resolved, one of five core passes, four of five visible passes, and zero hidden passes.

That split matters. The visible checks show the model was often in the neighborhood. The hidden oracle shows it did not close the job.

The recurring failures were not exotic: wrong or missing evidence, ignored-document ID errors, invoice classification mistakes, total calculation errors, warning-code problems, and proof-code failures.

The workflow run exposed the harder failure

The agentic workflow cases were worse.

In Workflow Case 04, the model produced some artifacts, but the final output was malformed and the run hit a token-limit/repair-loop pattern. The scored trace records `lmstudio_action_loop -> token_limit_stop`.

Cases 05, 06, and 07 failed around required artifacts and final output discipline. That is the practical problem: a local assistant that cannot reliably stop, write the required files, and leave a verifiable result is not doing the office work yet.

SVG did not rescue it

The City Plan SVG run also failed.

Gemma 4 12B Unified got one of three automated checks, but returned no parseable standalone SVG artifact. The run page therefore has no visual preview.

That is a clean failure. It is not a weak drawing. It is an absent artifact.

Practical readout

This is not a claim that Gemma 4 12B Unified is useless.

It is a claim about this workload. In the current Local Model Bench shape, the model can read pieces of the folder but fails the closure layer: exact JSON, evidence, hidden oracle, proof code, workflow artifacts, and constrained SVG output.

For private desktop work, that distinction is the point. A model that sounds close but leaves broken artifacts still leaves the human doing the cleanup.

Readout

Gemma 4 12B Unified looked partially capable at reading the folder, but not capable enough to finish it. In this benchmark, the difference between nearby and resolved is the whole result.