Search

What are you looking for?

Search our services, use cases and practical insights.

Enter at least 2 characters

Popular starting points

AI decisions

How should you compare AI models and providers?

Leaderboards show general capabilities, not whether a model supports your process reliably with your data. A defensible selection combines internal test cases with provider, operational and contractual criteria.

The short answer

First compare models on the same representative tasks and against explicit quality limits. Then assess provider terms, data flow, security, availability, integration, cost and portability. The best model is the option that reliably serves the specific purpose as part of the complete solution.

In brief

  • A public benchmark does not replace evaluation with the organisation's own cases and languages.
  • Model quality includes correctness, grounding, format, latency, robustness and cost.
  • Provider and model are separate decisions because the same model may be available through different deployment routes.
  • Selection criteria, test data, versions and accepted trade-offs should be documented for traceability.

Why does comparison start with the task?

Models perform differently across languages, document types, tool use and output formats. The general question 'Which model is best?' therefore has no dependable answer. First define what the application must do and which failures are unacceptable.

A test set should include typical cases, difficult exceptions, incomplete input and tasks the system should refuse or escalate. Every candidate receives the same application context and is scored with the same method.

  • Cover internal terminology and the languages used in practice
  • Include normal, edge, risk and unanswerable cases
  • Version model, prompt, sources and tool configuration
  • Set minimum thresholds and critical exclusions before testing

Which quality dimensions should be measured?

One average score can hide important weaknesses. Measure separately whether the solution is correct, complete, grounded and compliant with the required format. Production use also depends on latency, stability and behaviour when context is missing.

The case count, thresholds and weights below form one coherent document-extraction example and are not universal standards. Set them before testing to match the process, risk, language, volume and acceptable failures; the model weights in this example add up to 100 percent.

Which quality dimensions should be measured?
Model criterionMeasurementIllustrative minimumWeightEvidence
Domain task successBlind review of 180 representative casesAt least 97% correct mandatory fields; no critical failure35 %Versioned evaluation report with failure categories
Grounding and refusalCheck citations and deliberately unanswerable casesAt least 95% correct citations and 98% correct escalations25 %Evidence and refusal rationale for each test case
Format and tool useSchema validation and permitted tool paths99% valid format; no prohibited action15 %Automated test records and tool logs
RobustnessRepeat paraphrases, missing inputs and disruptionsQuality loss no greater than 5 percentage points10 %Regression suite across variants and model versions
Latency and costEnd-to-end measurement per successful case95th percentile below 8 seconds and under CHF 0.2015 %Measurements from an identical test environment

What should be assessed beyond the model?

The provider determines the regions and deployment modes in which a model is available, how input is processed and stored, and which service commitments apply. Support, change notices and access to logs also affect operations.

Keep model and provider assessments as two separate results. After mandatory gates have passed, the two weighted provider criteria in this example form their own 100-percent provider score; any later overall decision must show both scores and every exclusion explicitly.

What should be assessed beyond the model?
Provider criterionAssessmentGate or provider weightRequired evidence
Data termsMatch the exact service, plan, region, retention, training, deletion and subprocessorsMandatory gateContract, service description and documented data flow agree
Identity and securityTest SSO, roles, network, encryption, logs and incident notificationMandatory gateTechnical pilot plus current security and audit evidence
Operations and changeSimulate quotas, availability, model versions, retirement and a support case55% of the separate provider scoreSLA, change notice and recorded support test
Commercial and exitCalculate price tiers, commitments, liability, export and termination effects45% of the separate provider scoreThree-year cost model and successful export test

How does the selection remain traceable?

Weight the criteria before making a decision and ask business, IT, security, privacy and operations to define their minimum requirements. A candidate that violates an exclusion criterion should not be rescued by a high average score.

  • Use a shortlist of plausible candidates rather than an unfocused market survey
  • Score responses blindly or at least under identical conditions
  • Assess model, deployment route and provider separately
  • Retain a repeatable test for later model and version changes
Example from day-to-day business

Example: extracting data from technical service reports

Three models extract error code, component and action from 180 anonymised reports in German, French and English. The team measures field accuracy, correct blanks where information is absent, latency and cost. Two providers meet the quality bar. The final choice is the deployment route that supports existing identities, the required region and version control. The test set remains as a regression suite.

What to remember

Compare candidates on the same real tasks and explicit exclusion limits. Then decide on the complete deployment, not on a leaderboard position or model name.

Sources and further reading

These primary sources provide further detail on definitions, technical foundations or responsible use.

Content reviewed

Would you like to apply this to your situation?

Together, we clarify what makes sense for your process, data and systems – in plain language and without unnecessary complexity.

Discuss Your Project