[In preview] Public Preview: Evaluations with Intelligent Trace Sampling

Fournisseur: Microsoft
Produit: Microsoft Foundry
Type: Tarification
Date annonce: 2026-06-03
Date effet: Date non publiee
Impact: Concerné

En bref

Ce que dit la source

Microsoft Foundry observability introduces intelligent trace filtering and sampling for evaluations in public preview. Instead of running evaluations against all production traces, Foundry now select a representative subset of traces using a multi-stage MinHash farthest-first diversity algorithm, including deduplication, hard filers, aggregation and diversity-based selection for evaluation. This approach improves cost efficiency while producing significantly higher lexical diversity and diverse coverage compared to random sampling. Intelligent sampling is particularly effective for: evaluation and benchmarks, rubric generation and finetuning dataset curation. Learn more .

Pour continuer

Voir les autres changements de Microsoft Foundry Voir le prix public de Microsoft Foundry Comparer avec mon parc Poser une question

Lire la source primaire

Choisir votre lecture

Le rôle change l'angle de lecture, pas les faits, la date ni le niveau de preuve.

Lecture pour un commercial

Qui est concerné : The audience described by the Microsoft update is represented by: [In preview] Public Preview: Evaluations with Intelligent Trace Sampling Microsoft Foundry observability introduces intelligent trace filtering and sampling for evaluations in public preview. Instead of running evaluations against all production traces, Foundry now select a representative subset of traces using a multi-stage MinHash farthest-first diversity algorithm, including deduplication, hard filers, aggregation and diversity-based selection for evaluation. This approach improves cost effic

Pourquoi c'est important : The change matters because the official source describes: [In preview] Public Preview: Evaluations with Intelligent Trace Sampling Microsoft Foundry observability introduces intelligent trace filtering and sampling for evaluations in public preview. Instead of running evaluations against all production traces, Foundry now select a representative subset of traces using a multi-stage MinHash farthest-first diversity algorithm, including deduplication, hard filers, aggregation and diversity-based selection for evaluation. This approach improves cost effic

Urgence : Surveiller

Prochaine action : Review affected accounts with the customer using the official announcement.

Opportunités commerciales

Use the verified change to open a scoped customer conversation.

Actions techniques

Assess whether the documented change intersects the customer's current stack.

Questions à poser au client

Does this documented change affect a product or workload in scope?

Risques et objections

The official source does not establish facts beyond the quoted material.

Points à confirmer

The effective date is unknown and must be confirmed before scheduling action.
The affected customer scope is unknown and must be confirmed before action.

Preuves et traçabilité

Chaque extrait est relié à la source primaire et conservé pour vérification.

Capture brute
2026-08-30T06:15:07.776356+00:00
Événement
[In preview] Public Preview: Evaluations with Intelligent Trace Sampling Microsoft Foundry observability introduces intelligent trace filtering and sampling for evaluations in public preview. Instead of running evaluations against all production traces, Foundry now select a representative subset of traces using a multi-stage MinHash farthest-first diversity algorithm, including deduplication, hard filers, aggregation and diversity-based selection for evaluation. This approach improves cost efficiency while producing significantly higher lexical diversity and diverse coverage compared to random sampling. Intelligent sampling is particularly effective for: evaluation and benchmarks, rubric generation and finetuning dataset curation. Learn more .
source
Produit Microsoft Foundry
[In preview] Public Preview: Evaluations with Intelligent Trace Sampling Microsoft Foundry observability introduces intelligent trace filtering and sampling for evaluations in public preview. Instead of running evaluations against all production traces, Foundry now select a representative subset of traces using a multi-stage MinHash farthest-first diversity algorithm, including deduplication, hard filers, aggregation and diversity-based selection for evaluation. This approach improves cost efficiency while producing significantly higher lexical diversity and diverse coverage compared to random sampling. Intelligent sampling is particularly effective for: evaluation and benchmarks, rubric generation and finetuning dataset curation. Learn more .
source
Insight un commercial
[In preview] Public Preview: Evaluations with Intelligent Trace Sampling Microsoft Foundry observability introduces intelligent trace filtering and sampling for evaluations in public preview. Instead of running evaluations against all production traces, Foundry now select a representative subset of traces using a multi-stage MinHash farthest-first diversity algorithm, including deduplication, hard filers, aggregation and diversity-based selection for evaluation. This approach improves cost efficiency while producing significantly higher lexical diversity and diverse coverage compared to random sampling. Intelligent sampling is particularly effective for: evaluation and benchmarks, rubric generation and finetuning dataset curation. Learn more .
source

Retour au fil public