LLM·Dex
Newsletter
Use caseTop 3 picks

Best Realtime Voice Models in 2026

Low-latency speech-in / speech-out voice assistants.

Updated

How we ranked

  • End-to-end latency (mic on to first word out)
  • Interruption handling
  • Voice naturalness
  • Tool-use mid-conversation
  • Cost per minute

Read the full methodology for our sourcing and ranking standards.

Realtime voice is the killer demo of the multimodal era. The model has to listen, transcribe, reason, and respond as fast as a human conversation tolerates, typically under 800ms end-to-end.

OpenAI's Realtime API (powered by GPT-5/-mini) and Gemini Live are the two clearly-deployable options. Anthropic doesn't ship a native realtime voice product yet. Open-weight stacks (Pipecat + Cartesia + Llama) are competitive but require integration work.

If you're shipping a voice agent today, OpenAI Realtime is the lowest-friction path. If cost matters more than maximum quality, Gemini Live with Flash is dramatically cheaper.

The ranking

  1. #1OpenAI

    GPT-5

    OpenAI's unified flagship combining GPT-line breadth with built-in reasoning, replacing both GPT-4o and the o-series for most users.

    Full specsTry GPT-5
    Context
    400K tokens
    Output · 1M
    $10.00 / 1M tokens
    Modalities
    text, vision, audio

    Why it ranks here. Unified model, reasoning routed automatically per query. Excellent tool-use and JSON-mode discipline. Tracked weakness: Reasoning routing means latency is unpredictable per query.

  2. #2Google

    Gemini 3 Flash

    Google's high-speed, low-cost mid-tier with the same massive context window, popular for high-volume RAG.

    Full specsTry Gemini 3 Flash
    Context
    1.0M tokens
    Output · 1M
    Pricing not published
    Modalities
    text, vision, audio, video

    Why it ranks here. 1M-token context at mid-tier price. Very fast, good for interactive UX. Tracked weakness: Reasoning quality below Pro.

  3. #3OpenAI

    GPT-5 mini

    GPT-5's mid-tier sibling, most of the quality at a fraction of the price, ideal for high-volume production workloads.

    Full specsTry GPT-5 mini
    Context
    400K tokens
    Output · 1M
    $2.00 / 1M tokens
    Modalities
    text, vision, audio

    Why it ranks here. Excellent price-quality ratio for production workloads. Fast first-token latency. Tracked weakness: Quality gap vs. flagship visible on hard reasoning.

How to choose

Don't pick on the headline ranking alone. Run your top two picks on a representative sample of your own workload and compare. The numbers in this list are sound, but task-specific quality varies in ways no benchmark fully captures. The criteria above are the right axes to evaluate on, but the weighting depends on your stack.

  • Cost-sensitive workloads, start with the cheapest of the top three; escalate only if quality is the bottleneck.
  • Privacy-sensitive workloads, filter to open-weight picks above. They're labeled with a green badge.
  • Latency-sensitive workloads, see the Fastest LLMs list, which can override task-specific picks.

Frequently asked

  • What is the best model for realtime voice models?
    Our #1 pick is GPT-5 from OpenAI. OpenAI's unified flagship combining GPT-line breadth with built-in reasoning, replacing both GPT-4o and the o-series for most users.
  • How are these rankings determined?
    We rank by the criteria listed at the top of this page: End-to-end latency (mic on to first word out); Interruption handling; Voice naturalness. Where two models are close, we prefer the one with stronger production deployment evidence at the time of writing. Read the full methodology for our standards.
  • GPT-5 or Gemini 3 Flash?
    Both are top-tier picks. GPT-5 edges ahead on the criteria most relevant to this task. Gemini 3 Flash is the strongest alternative, see the head-to-head comparison page for full deltas.
  • Are open-source models on this list?
    Yes where they're competitive. Each entry below shows whether the model ships open weights and under what license.
  • How often is this list updated?
    Weekly. New launches that affect the ranking get reflected within seven days. The "last updated" stamp at the top of the page reflects the most recent dataset commit.

Related guides

Friday digest

The week's AI launches, in your inbox.

One short email every Friday, new models, leaks, and quietly-shipped APIs you missed.