Gemini 3.5 Transcribe Live costs ~$9/1k minutes, pricier than Cartesia and Deepgram
ArtificialAnlys · x · 2026-08-27
Artificial Analysis updated its ASR leaderboard with streaming and non-streaming results. Based on Google's token estimates and API pricing ($3.50/M input, $21/M output), Gemini 3.5 Transcribe Live runs about $9 per 1,000 minutes — above Cartesia Ink-2 ($4), Deepgram Nova-3 Realtime ($4.80), and ElevenLabs Scribe v2 Realtime ($6.50).
The leaderboard uses its AA-WER v2 metric: a duration-weighted average over 8 hours across AA-AgentTalk (50%), VoxPopuli, and Earnings22, alongside speed factor and price comparisons.
Related event: Gemini 3.5 Transcribe Live is faster and more accurate but pricier(3 posts)→
More from Models
- Z.AI Unveils GLM-5.3-Flash: 320B Native Multimodal Model with 90% Cost Cut — baseten · 2026-08-27
- MiniMax H3 Max tops video leaderboards via fal's post-training — ArtificialAnlys · 2026-08-27
- GLM-5.3 Flash Review: GPT-5.6 Level Performance at Ultra-Low Cost — zainhas · 2026-08-27
- Qwen 3.8 and GLM 5.3 Flash open models released, available for Dell on-prem deployment — _akhaliq · 2026-08-27
- Fal secures new funding as H3 Max model achieves fastest speed and best quality — isidentical · 2026-08-27
- Local model choice in 2026: Qwopus 9B vs Qwen 27B — Ammargok · 2026-08-27