Snap 2

Write a Quorlyn product-news email for people who use the Quorlyn API. Cover the facts below.

Quorlyn Snap 2.0 is Quorlyn's most capable speech-to-speech voice model, with improved intelligence, transcription accuracy, and conversational capabilities. Try it at https://quorlyn.s3.amazonaws.com/console/voice/agents

Voice docs: https://quorlyn.s3.amazonaws.com/docs/voice/speech-to-speech

More accurate transcription: 1.5–2× better word error rate than dedicated STT models across 24 languages. Up to ~10× better in noisy, telephony conditions.

Efficient parallel reasoning: Reasons while speaking with ~60% fewer reasoning tokens than 1.0, so tool calls and responses feel snappier.

More natural conversation: Shorter sentences, one question at a time, less filler—trained to sound like real conversation while still driving complex workflows.

On August 5, 2026, quorlyn-voice-latest will move from snap-1.0 to snap-2.0. No action needed to upgrade. To stay on 1.0, pin snap-1.0 before then.

Priced at $0.08/min of audio. Full details in the announcement at https://quorlyn.s3.amazonaws.com/news/snap-2 and the pricing docs at https://quorlyn.s3.amazonaws.com/docs/pricing

For the artwork, use assets.illustrations.snapSpeechChart, a speech-to-speech model card for Quorlyn Snap 2.0 with three indigo stat tiles: 1.5–2× word error rate versus dedicated STT, ~60% fewer reasoning tokens than 1.0, and $0.08 per minute of audio.

Use /world/quorlyn-email for brand chrome and CSS. Use the absolute HTTPS URLs from tokens.json; do not use /world/ paths as image src.

Dark mode should be supported.

Leave the HTML in output/email.html, a plain-text version in output/email.txt, and the subject in output/meta.json as {"subject": "..."}. Don't include From, To, or Date headers.