ONE WORKSPACE. TWO DIRECTIONS.

Your documents.
Your choice.
Cloud or local.

Check an agreement. Explain a charge. Find what is still missing. AMBER connects your question with the relevant document passages; the model helps turn them into an answer you can review.

Choose where the model works.

Start with the way your team works, the information it handles and the equipment it has.

Gemini · cloud

For a responsive working conversation.

The answer model runs with the provider. Your team uses AMBER in a browser, without a local machine sized to run the model.

Selected document passages are sent to the provider to prepare the answer.

Observed response times
2–4 seconds for simple requests; 17–36 seconds for more involved questions.
What to budget for
Model usage charges, application hosting and maintenance.

Qwen · local

For analysis on your own equipment.

Our trials show that a local model can work with agreements, AML materials and legal documents. This route suits teams that want the answer model on their own infrastructure.

The local route was tested in research. We agree the complete data flow and prepare the deployment for each environment.

Observed response times
Around 70 seconds typically in our public-document trial.
What to budget for
Suitable hardware, electricity and maintenance, alongside the application.

Times come from different studies and tasks. They illustrate the experience, not a direct speed comparison or a delivery promise.

What we observed.

Six document tasks. The same supplied evidence. Five local models, with Gemini as a cloud reference.

Requirements met in the six-task comparison
Gemini 3.8 Flash high34 / 34
Qwen 3.6-35B-A3B Q430 / 34
Ornith 1.5-35B-A3B Q429 / 34
Gemma 4-26B-A4B Q426 / 34
Nemotron 3.5 Lightning Q419 / 34
GLM 4.7 Flash Q412 / 34

34 individual checks across six prepared tasks. These scores are not the percentage of fully correct answers.

Gemini met all 34 checks in this comparison.

Qwen met 30. It handled useful calculations and interpretation, but needed closer review of sources and answer completeness.

Check important conclusions and their sources with every model.

About these observations

Start with a useful question.

01

Check the terms

Which agreement applies, and what changed in the amendment?

02

Explain a charge

What remains payable after the credit already applied?

03

Review the evidence

What do these ownership records establish, and what is still missing?

About these observations

Research observations · September 2026

The six-task comparison used prepared documents and the same drafting instructions. Gemini 3.8 Flash high met 34/34 checks; Qwen3.6-35B-A3B, with thinking off, met 30/34. Model settings differed. The documents were supplied directly: this comparison measures the answers, not a gain attributable to AMBER search.

Task-by-task results
ModelRequirements metComplete casesMedian timeWhat to consider
Gemini 3.8 Flash highCloud reference34 / 346 / 6Not comparableMet all measured requirements in this sample; important conclusions still need review.
Qwen 3.6-35B-A3B Q430 / 343 / 656.3 sHighest local score in this sample; omitted required details and the historical comparison.
Ornith 1.5-35B-A3B Q429 / 342 / 635.9 sClose to Qwen, with shorter responses; double-counted a credit and overstated one ownership share.
Gemma 4-26B-A4B Q426 / 341 / 622.5 sShortest local median here; double-counted a credit and omitted required distinctions.
Nemotron 3.5 Lightning Q419 / 340 / 629.5 sErrors in amounts, party identification and source references.
GLM 4.7 Flash Q412 / 340 / 623.0 s · attemptsFive final answers delivered; numerical errors and one answer cut off at the output limit.

A complete case meets every task requirement and the application’s source-reference and comparison rules. Zero complete cases does not mean every statement was wrong. These six prepared tasks are observations, not a general model ranking.

Local medians cover drafting plus source checking on the same six supplied contexts, excluding retrieval, import and scan processing. GLM delivered five of six final answers; its 23.0-second median includes the failed 194.9-second attempt. The separate Vertex times above are not timings for this six-task Gemini result.

Primary, strict scores are shown. Alternative interpretations of individual requirements give Ornith 30/34 and 3/6 complete cases, Nemotron 20/34, or GLM 13/34. These alternatives do not replace the primary assessment.

Where each model succeeded or stumbled
TaskGemini 3.8 Flash highQwen 3.6-35B-A3B Q4Ornith 1.5-35B-A3B Q4Gemma 4-26B-A4B Q4Nemotron 3.5 Lightning Q4GLM 4.7 Flash Q4
Freight charge correction6/6All six checks met5/6Correct balance; unsupported claim about the rate’s origin4/6596 instead of 620: credit deducted twice4/6596 instead of 620: credit deducted twice1/6Free waiting time omitted0/6266 instead of 140 for waiting
Earlier contract terms6/6All six checks met5/6Correct terms; required comparison not accepted5/6Correct terms; required comparison missing6/6All six checks met4/6Retroactive effect denied despite the source5/6Comparison order reversed
Laboratory release5/5All five checks met5/5All five checks met5/5All five checks met3/5Other instrument and lot not explicitly excluded3/5Same omissions; English instead of German3/5Other instrument and lot not explicitly excluded
Ownership and control6/6All six checks met6/6All six checks met5/672% instead of 36% for one owner5/6No explicit request for full identifiers6/6Six criteria met; source declaration missing2/625.2% instead of 36%; other shares omitted
Pleading review6/6All six checks met4/6A2 section 2 omitted; supplied court rejected5/6Court identified, but source and missing-field distinction confused4/6A2 section 2 and missing attachment link omitted3/6Wrong company and incomplete clause reference0/6No final answer: output limit reached
Spend and coverage5/5All five checks met5/5All five checks met5/5All five checks met4/5Earlier-record completeness caveat omitted2/531,600 instead of 29,600: credit not subtracted2/522,000 instead of 29,600

Scores count individual requirements. Notes give selected examples; a full score does not guarantee flawless wording or acceptance by every application check.

Cloud times cover eight successful simple or conversational requests (2.12–3.53 seconds) and four complex requests (16.84–36.05 seconds) through Google Vertex on small prepared document sets. Local times come from 18 Polish questions over eight public AML, customs and legal documents: median 69.77 seconds, range 48.28–96.73 seconds. These are separate studies and application versions; import and scan processing are excluded. Concurrent use was not measured.

In the public-document study, Qwen met 50/54 content checks; 10/18 answers met all content, language and source-support requirements (11/18 under an independent review). The trial did not meet its pre-set quality and response-time targets. Local use therefore remains a route to assess in a pilot, with human review.

Try the question that matters to you.

Bring a small, agreed sample of documents. Together we assess the answer, its sources and the time it takes before choosing the deployment.

Discuss your documents

Public reference materials were used for research. AMBER is not affiliated with the organizations named in those materials.