← Back to all models
Latest speech recognition model with improved accuracy across 100+ languages, real-time streaming, and speaker diarization.
#speech-to-text#multilingual#open-weight#diarization
🧮
2B
Parameters
📏
N/A
Context Window
🔓
Open Source
License
📅
Oct 1, 2025
Released
💰 Pricing
Input
$0.006/min
per 1M tokens
Output
N/A
per 1M tokens
⚡ Strengths
✓100+ languages
✓Speaker diarization
✓Real-time streaming
✓Open weights
🎯 Use Cases
TranscriptionMeeting notesSubtitlesVoice apps
🔌
API Available
This model is accessible via API for integration into your applications.
⭐ Related Models
🌐
GPT-5
OpenAI
🔥 96
OpenAI's latest flagship. Advanced multimodal capabilities with native tool use, improved reasoning, and massive knowledge.
🔒 proprietary📏 256K ctx💰 $10
Dec 1, 2025View details →
🌐
GPT-4o
OpenAI
🔥 82
Omni model supporting text, vision, and audio natively. Fast and capable with strong multimodal understanding.
🔒 proprietary📏 128K ctx💰 $2.50
May 13, 2024View details →
💬
GPT-4o Mini
OpenAI
🔥 70
Compact and affordable. Surprisingly capable for its price point, ideal for high-volume applications.
🔒 proprietary📏 128K ctx💰 $0.15
Jul 18, 2024View details →