Observed Signal · Sep 24, 2026 · Product Launch · Source: https://martechseries.com/feed/ · Impact: 2/5 · Sentiment: Positive

Guava Launches Daytona Voice Model and Open Benchmark

Executive Signal Summary

Guava released Daytona, a new voice model, and introduced the Guava Voice Index, an open scoring framework for voice agent quality. The index evaluates agents on responsiveness, conversational flow, fidelity, resolution, and TTS quality, with content accuracy weighted at 60% and voice experience at 40%. It combines automated metrics from EVA-Bench and Coval's benchmark with blind pairwise human evaluations. In September 2026 tests, Daytona scored highest at 58.88, ahead of ElevenLabs with GPT 5.4, Grok, and ElevenLabs with Haiku 4.5. However, no system exceeded 60, and human agents outperformed all AI systems in response timing and interruption recovery. Guava operates the full model stack and trains on over 10 billion minutes of live conversations. The full methodology is published publicly.

Polaris7 AgentPolaris7 Strategic Assessment
High Confidence

The launch of a new voice model and an open benchmark contributes to AI capabilities relevant to customer experience, but it is not a major platform shift.

SIGNAL RADAR

Track ElevenLabs Signals & Market Shifts in Real-Time

Polaris7 autonomous intelligence agents track regulatory filings, primary sources, executive changes, and deal flow 24/7. Create your Polaris7 workspace to monitor these entities.

Start Monitoring
Verified primary sourcesAutomated briefingsInstant watchlist setup

Key Takeaways & Evidence Grounding

  • Guava launched Daytona, a new voice model, and the Guava Voice Index (GVI), an open benchmark.
  • Daytona scored 58.88 out of 100, ranking first among tested systems.
  • The GVI evaluates voice agents on five pillars: responsiveness, conversational flow, fidelity, resolution, and TTS quality.
  • Humans outperformed all AI systems: preferred in 91% of response timing and 93% of interruption recovery comparisons.
  • Guava's models are trained on over 10 billion minutes of live agent conversations since 2013.

Connected Companies & Entities

“ElevenLabs paired with GPT 5.4 and Haiku 4.5 were among the systems tested.”

“The GVI draws on components of EVA-Bench, the voice agent evaluation suite published by ServiceNow.”

Ontology Mapping & Concepts

Large Language Models (LLM) & AIConversational AI & Chatbots#Guava Voice Index (GVI)#Voice agent benchmark#Human-agent comparative evaluation
Primary Source Grounding & Direct Attribution
Direct Origin Attribution
Primary Reporting: https://martechseries.com/feed/•Published: Sep 24, 2026
Original Coverage Title: “Guava Launches Daytona, a State of the Art Voice Model, alongside an Open Benchmark Evaluating Voice Agent Performance”

Related Market Signals & Shifts

Recent verified developments and strategic activity across this market segment.

AI & Machine LearningSep 16, 2026

Google Launches Android Bench 2.0 with Long-Horizon Tasks, Agentic Evaluation

Google announced the release of Android Bench 2.0, an upgraded benchmark for evaluating large language models (LLMs) and coding agents on Android development tasks. The update introduces long-horizon tasks (LHTs) that simulate complex, multi-day engineering work, moving beyond simple bug fixes. Android Bench 2.0 also incorporates continuous scoring instead of binary pass/fail, providing a more nuanced assessment of model performance. Additionally, Google has begun agentic evaluations, pairing models with agents like GPT-5.6 Sol on Codex and Gemini 3.8 Flash on Google Antigravity to measure real-world workflow effectiveness. The leaderboard has been expanded with new models, including Gemini 3.8 Flash, OpenAI GPT-6, and Anthropic Fable 5.1, with OpenAI GPT-6 Astra achieving the highest pass rate of 28% on LHTs, compared to 91% on original tasks. This release reflects Google's commitment to improving AI-assisted Android development through more realistic and rigorous evaluation.

Read assessment
Conversational AI & ChatbotsAug 26, 2026

3CLogic Launches AI Agent Evaluator for Voice AI QA

3CLogic announced the release of AI Agent Evaluator, an automated quality assurance (QA) and scoring engine integrated into its Voice AI Hub. The feature automatically scores 100% of Voice AI interactions against configurable metrics such as resolution, goal completion, and quality, and provides plain-language reasoning for each rating without manual transcript review. Built to address scaling limits of manual QA and tools designed for human agents, the evaluator offers role-specific yardsticks, automated task validation (verifying system actions like ticket updates), and real-time dashboards for performance insights. The company says the capability lets organizations hold AI agents to the same standards as live agents and is now generally available to Voice AI Hub customers.

Read assessment
Conversational AI & ChatbotsMay 7, 2026

OpenAI launches three Realtime voice models

OpenAI announced the addition of three realtime voice-intelligence models to its Realtime API on May 7, 2026: GPT‑Realtime‑2, a GPT‑5‑class reasoning voice model for realistic conversational and agentic workflows; GPT‑Realtime‑Translate, which provides live translation with support for more than 70 input languages and 13 output languages; and GPT‑Realtime‑Whisper, a low-latency streaming speech-to-text transcription capability. The features are intended for customer service, education, media, events and creator platforms. Translate and Whisper are billed by the minute while GPT‑Realtime‑2 is billed by token consumption. OpenAI said it has embedded safety guardrails and active classifiers to halt conversations that violate harmful-content policies. The announcement follows OpenAI’s published evaluations showing gains versus prior realtime models and details pricing and safety controls in the Realtime API documentation.

Read assessment
Executive Briefing & Watchlist

Receive all future signals for ElevenLabs & competitors in your briefing

Never miss material M&A deals, leadership changes, account wins, or product rollouts. Polaris7 monitors 20,000+ primary sources 24/7, delivering verified telemetry directly into your executive briefing.

Daily primary source scansCurated executive briefingsUnsubscribe anytime