AI Index

Realtime voice models

Realtime voice models: speech in and speech out over a live session, as the conversation goes. Voice chat models answered one request at a time are language models. Prices per 1M audio tokens on the provider’s own API where it prices by tokens; per minute of audio where it prices by time. Rows show standard prices; other price tiers are on each model’s page. Dots show where each value came from; a hollow dot means it is not verified yet.

38 of 45 realtime voice models, 6 dated snapshots under their models

  • StepAudio 3 Realtime Verified
    StepFunOct 31
    —
    Context Not stated
    Audio in / 1M —
    Audio out / 1M —
    Per minute $0.125
  • qwen-audio-3.1-realtime-flash Verified
    Alibaba QwenOct 27
    Tool use
    Context 262.14K
    Audio in / 1M $0.93
    Audio out / 1M $1.87
    Per minute —
  • Eleven Realtime v1 Max Verified
    ElevenLabsOct 20
    —
    Tool use
    Context Not stated
    Audio in / 1M —
    Audio out / 1M —
    Per minute $0.080
  • Eleven Realtime v1 Mini Verified
    ElevenLabsOct 20
    Context Not stated
    Audio in / 1M —
    Audio out / 1M —
    Per minute $0.080
  • Grok Voice 3.0 Partly verified
    xAIOct 14
    Context Not stated
    Audio in / 1M —
    Audio out / 1M —
    Per minute —
  • Nova 2.5 Sonic Verified
    Amazon Web ServicesOct 5
    —
    Tool use
    Context 256K
    Audio in / 1M $3.00
    Audio out / 1M $12.00
    Per minute —
  • Qwen3.8-Omni-Flash-Realtime Verified
    Alibaba QwenSep 21
    Image inputVideo input
    Tool use
    Context 196.61K
    Audio in / 1M $0.93
    Audio out / 1M $1.87
    Per minute —
  • qwen-audio-3.1-realtime-plus Verified
    Alibaba QwenSep 20
    Tool use
    Context 262.14K
    Audio in / 1M $6.40
    Audio out / 1M $24.00
    Per minute —
  • Gemini 3.8 Live Verified
    GoogleSep 15
    Image input
    ReasoningTool use
    Context 131.07K
    Audio in / 1M $3.00
    Audio out / 1M $12.00
    Per minute $0.018
  • Gemini 3.8 Live Extended Thinking Verified
    GoogleSep 15
    Image input
    ReasoningTool use
    Context 131.07K
    Audio in / 1M $3.00
    Audio out / 1M $12.00
    Per minute $0.018
  • GPT-Live 1 Verified
    OpenAISep 10
    ReasoningTool use
    Context Not stated
    Audio in / 1M —
    Audio out / 1M —
    Per minute $0.050
  • Grok Voice 2.0 Partly verified
    xAIJul 23
    —
    Context Not stated
    Audio in / 1M —
    Audio out / 1M —
    Per minute —
  • qwen-audio-3.0-realtime-flash Verified
    Alibaba QwenJul 14
    ReasoningTool use
    Context 40.96K
    Audio in / 1M $0.93
    Audio out / 1M $1.87
    Per minute —
  • qwen-audio-3.0-realtime-plus Verified
    Alibaba QwenJul 14
    Tool use
    Context 40.96K
    Audio in / 1M $6.40
    Audio out / 1M $24.00
    Per minute —
  • GPT-Realtime-2.1 Partly verified
    OpenAIJul 6
    Image input
    ReasoningTool use
    Context 128K
    Audio in / 1M $32.00
    Audio out / 1M $64.00
    Per minute —
  • GPT-Realtime-2.1 Mini Verified
    OpenAIJul 6
    Image input
    ReasoningTool use
    Context 128K
    Audio in / 1M $10.00
    Audio out / 1M $20.00
    Per minute —
  • GPT-Realtime-2 Verified
    OpenAIMay 7
    Image input
    ReasoningTool use
    Context 128K
    Audio in / 1M $32.00
    Audio out / 1M $64.00
    Per minute —
  • Gemini 3.1 Flash Live preview Partly verified
    GoogleMar 26 Deprecated
    Image input
    ReasoningTool use
    Context 131.07K
    Audio in / 1M $3.00
    Audio out / 1M $12.00
    Per minute $0.018
  • Qwen3.5-Omni-Flash-Realtime Verified
    Alibaba QwenMar 26
    Image inputVideo input
    Tool use
    Context 262.14K
    Audio in / 1M $4.50
    Audio out / 1M $17.70
    Per minute —
  • qwen3.5-omni-plus-realtime Verified
    Alibaba QwenMar 26
    Image inputVideo input
    Tool use
    Context 262.14K
    Audio in / 1M $16.50
    Audio out / 1M $62.00
    Per minute —
  • GPT-Realtime-1.5 Verified
    OpenAIFeb 23
    Image input
    Tool use
    Context 32K
    Audio in / 1M $32.00
    Audio out / 1M $64.00
    Per minute —
  • Gemini 2.5 Flash Native Audio Verified
    GoogleDec 12, 2025
    Video input
    ReasoningTool use
    Context 131.07K
    Audio in / 1M $3.00
    Audio out / 1M $12.00
    Per minute —
  • Qwen3-Omni Flash Realtime Partly verified
    Alibaba QwenDec 4, 2025
    Image inputVideo input
    Context 65.54K
    Audio in / 1M —
    Audio out / 1M —
    Per minute —
  • Nova 2 Sonic Verified
    Amazon Web ServicesDec 2, 2025
    ReasoningTool use
    Context 1M
    Audio in / 1M $3.00
    Audio out / 1M $12.00
    Per minute —
  • GPT-Realtime Mini Verified
    OpenAIOct 6, 2025 Deprecated
    Image input
    Tool use
    Context 32K
    Audio in / 1M $10.00
    Audio out / 1M $20.00
    Per minute —
  • GPT-Realtime Verified
    OpenAIAug 28, 2025 Deprecated
    Image input
    Tool use
    Context 32K
    Audio in / 1M $32.00
    Audio out / 1M $64.00
    Per minute —
  • qwen-omni-turbo-realtime Partly verified
    Alibaba QwenMay 8, 2025
    Image inputVideo input
    Context 32.77K
    Audio in / 1M —
    Audio out / 1M —
    Per minute —
  • GPT-4o Realtime Partly verified
    OpenAIDec 17, 2024 Deprecated
    Context 32K
    Audio in / 1M $5.00
    Audio out / 1M $20.00
    Per minute Not stated
  • GLM-Realtime-Air Verified
    Zhipu AI
    Video input
    Context —
    Audio in / 1M —
    Audio out / 1M —
    Per minute ¥0.30
  • GLM-Realtime-Flash Verified
    Zhipu AI
    Video input
    Context —
    Audio in / 1M —
    Audio out / 1M —
    Per minute ¥0.18
  • GPT-4o Mini Realtime Partly verified
    OpenAI Deprecated
    Context 16K
    Audio in / 1M $0.60
    Audio out / 1M $2.40
    Per minute Not stated
  • StepAudio 2.5 Realtime Verified
    StepFun
    Context Not stated
    Audio in / 1M $1.50
    Audio out / 1M $10.00
    Per minute —

38 of 38 models expanded