nafyi.

Customer testimonials published by Anthropic for Claude Sonnet 5.5

Anthropic News ·

Anthropic’s Claude Sonnet 5.5 announcement includes named customer testimonials from Epic Games, Base44, Slack and Zendesk. These customers describe their own engineering, app-building, assistant and support tests. The accounts are selected and published by Anthropic; their reported results have not been independently verified by nafyi. Lisez 4 points de vue avec leurs éléments à l’appui et les liens vers les sources.

Daniel VogelGabriel GrinbergCurtis AllenAbhinay Kathuria
Lire les points clésOuvrir Éditeur (s’ouvre dans un nouvel onglet)

En un coup d’œil

  • Epic reports higher-tier quality on engineering tasks

    In a customer testimonial published by Anthropic, Epic Games’ Daniel Vogel says Sonnet 5.5 met the quality bar he expected from a higher-tier model in early system-design and data-flow reviews, while handling long engineering tasks with less prescriptive prompting.

    Lire le moment probant · Source text block 30 · customer quotation

  • Base44 reports fewer iterations across 118 app builds

    In a customer testimonial published by Anthropic, Base44’s Gabriel Grinberg reports that Sonnet 5.5 matched Opus 5’s scores across 118 app builds, averaging 3.6 iterations per build versus 7.7 for Opus 5.

    Lire le moment probant · Source text block 34 · customer quotation

  • Slack reports better offline results with fewer tokens

    In a customer testimonial published by Anthropic, Slack’s Curtis Allen reports that Sonnet 5.5 outperformed Sonnet 5 on almost all of their offline Slackbot evaluations without prompt changes, using fewer steps and about 14% fewer output tokens.

    Lire le moment probant · Source text block 41 · customer quotation

  • Zendesk reports fewer wrong decisions and faster processing

    In a customer testimonial published by Anthropic, Zendesk’s Abhinay Kathuria says Sonnet 5.5 made fewer wrong decisions on hundreds of support use cases and processed tickets 20% faster than the Claude models Zendesk was using at the time.

    Lire le moment probant · Source text block 42 · customer quotation

Passages clés4

Passages attribués et accompagnés du contexte nécessaire à leur vérification. Ouvrez le texte original pour vérifier la source.

Source text block 34 · customer quotation (s’ouvre dans un nouvel onglet)
Application building iterations

Base44 reports fewer iterations across 118 app builds

Extrait original

“Across 118 real app builds, Claude Sonnet 5.5 produced apps that scored level with Opus 5. It got there in 3.6 iterations per build on average, where Opus 5 took 7.7. It had the fewest failed tool calls of any model we compared. It also rarely stopped mid-build to ask the user a question, so fewer builds stall waiting on someone to answer.”
Gabriel Grinberg·
Source text block 41 · customer quotation (s’ouvre dans un nouvel onglet)
Assistant evaluation efficiency

Slack reports better offline results with fewer tokens

Extrait original

“Without changing any of our prompts, Claude Sonnet 5.5 did better than Sonnet 5 on almost all of our offline Slackbot evals, in fewer steps and with about 14% fewer output tokens. When someone gives Slackbot a task, quality and speed are what matter most, and Sonnet 5.5 allows Slackbot to deliver better outcomes for users, faster.”
Curtis Allen·
Source text block 42 · customer quotation (s’ouvre dans un nouvel onglet)
Customer support workflows

Zendesk reports fewer wrong decisions and faster processing

Extrait original

“We fed Claude Sonnet 5.5 hundreds of real support use cases across replies and escalation requests. It made fewer wrong decisions and resolved tickets faster than the Claude models we use in production today. Tickets were processed 20% faster, getting our customers the help they need without the wait.”
Abhinay Kathuria·
Source text block 30 · customer quotation (s’ouvre dans un nouvel onglet)
Coding task quality

Epic reports higher-tier quality on engineering tasks

Extrait original

“In Epic’s early testing, Claude Sonnet 5.5 cleared the same quality bar you’d expect from a higher-tier model, holding up on a system design audit and a data flow review. The new model managed tens of thousands of lines of code for gameplay system architecture, kept responses snappy, handled multi-hour tasks, and delivered with less prescriptive prompting.”
Daniel Vogel·

Source et méthodologie

Ces points de vue renvoient à leurs sources originales. Les reformulations sont signalées et ne sont pas des citations mot à mot.

Ouvrir la transcription ou les documents sources (s’ouvre dans un nouvel onglet)Signaler un problème