█

AI/TLDR

Buy me a coffee
Buy me a coffee

Mercury Voice

Inception's diffusion language model tuned for voice agents, generally available to enterprise customers from September 29, 2026 — 320 ms median time to first answer token on real customer-service prompts.

IN SHORT

What is Mercury Voice?

Mercury Voice is Inception's diffusion language model tuned to power voice agents, made generally available to enterprise customers on 29 September 2026 after a preview announced alongside Mercury 2.5. It has a 128K-token context window and costs $0.40 per million input tokens and $1.50 per million output tokens. You can use it through Inception API (enterprise).

Where to use it
Inception API (enterprise)
Updated Checked against 3 sourcesMarkdown version ↗
Cite this pageAI/TLDR, "Mercury Voice: specs, pricing and benchmarks", https://ai-tldr.dev/models/mercury-voice/ (updated 1 October 2026)
// FACTSas of 1 October 2026
Context window128Kmax output 50K[1]
Input price$0.40 / 1M tokenscached $0.04[2]
Output price$1.50 / 1M tokens[2]
LicenceProprietary · API only[1]
Released29 September 2026[3]
  1. [1] inceptionlabs.ai Inception Labs
  2. [2] docs.inceptionlabs.ai Models & pricing (Inception docs)
  3. [3] inceptionlabs.ai Introducing Mercury Voice (Inception blog)
Mercury VoiceAPI onlyGenerally available to enterprise customers through the OpenAI-compatible Inception API; access is enabled per organization by Inception's sales team. Previewed alongside Mercury 2.5 in September 2026.
Released
29 Sep 2026
Context
128K
Input
$0.40 / 1M tokens
License
Proprietary

Overview

Mercury Voice is Inception's diffusion language model tuned to power voice agents, made generally available to enterprise customers on 29 September 2026 after a preview announced alongside Mercury 2.5. It is a text model that sits in the LLM slot of a voice pipeline: it reasons, calls tools, returns structured outputs and follows long system prompts, served through an OpenAI-compatible chat-completions endpoint so it plugs into LiveKit, Pipecat, Vapi, Retell or a custom voice stack.

Inception's metric for the release is time to first answer token (TTFAT) — how long until the model starts producing the words the caller hears, after all reasoning is done — measured against a conversational budget of about 500 ms. On OpenCall customer-service prompts, Inception reports 320 ms median (p50) and 750 ms p95 TTFAT at low reasoning effort, which it describes as 5.9× faster than GPT-6 Luna (no reasoning) and the only model in its test with a median under the 500 ms budget.

On quality, Inception publishes an average across τ³-bench Telecom, Retail and Airline, IFBench and BFCL v4 multi-turn: 70.5 for Mercury Voice at low effort, against 59.0 for Gemma 4 31B (non-reasoning) and 51.4 for GPT-4.1, the non-reasoning models it names as the usual defaults for voice agents. These are Inception's own figures from the launch post; no independent evaluation is cited.

Pricing is $0.40 per million input tokens and $1.50 per million output tokens, with a 50% launch discount to $0.20 / $0.75. Inception estimates that this works out to about $0.009 per minute of conversation for a typical voice-agent profile, compared with $0.045 per minute for GPT-4.1. Weights are not published.

Released2026-09-29
LicenseProprietary
WeightsAPI only
Context128K
Max output50K
ArchitectureDiffusion language model (dLLM) tuned for voice agents, with low, medium and high reasoning-effort settings
ModalitiesText
StatusGenerally available to enterprise customers through the OpenAI-compatible Inception API; access is enabled per organization by Inception's sales team. Previewed alongside Mercury 2.5 in September 2026.

Benchmarks

Bar chart of time to first answer token on OpenCall customer-service prompts, p50 and p95, for Mercury Voice at low, medium and high effort against Cerebras GPT OSS 120B, GPT-4.1, Gemini 3.5 Flash Lite, GLM-5.3 Flash, Gemma 4 31B, GPT-6 Luna, GPT-5.6 Luna, Qwen 3.5 397B and Claude Haiku 4.5, with a ~500 ms conversational budget line
Time to first answer token on real voice-agent prompts (OpenCall), p50 and p95 — Inception
Bar chart of average quality scores: Mercury Voice low 70.5, high 69.9 and medium 69 against GLM-5.3 Flash, Gemma 4 31B, Qwen 3.5 397B, GPT-6 Luna, Cerebras GPT OSS 120B, Gemini 3.5 Flash Lite, GPT-5.6 Luna, Claude Haiku 4.5 and GPT-4.1
Average quality across τ³-bench Telecom, Retail, Airline, IFBench and BFCL v4 — Inception
Scatter plot of average τ³-bench quality against average p50 time to first answer token on a log scale, with Mercury Voice (low) the only model inside the ~500 ms conversational budget
τ³-bench quality vs. latency (mean of Telecom, Retail and Airline) — Inception

Mercury Voice (low effort) vs the non-reasoning models Inception names as voice-agent defaults

BenchmarkMercury Voice (Low)Gemma 4 31B (non-reasoning)GPT-4.1
OpenCall TTFAT p50 / p95320 ms / 750 ms1.24 s / 3.51 s870 ms / 1.45 s
Average quality70.55951.4
τ³-bench Telecom77.25048.2
τ³-bench Airline786654
τ³-bench Retail69.369.364
IFBench67.454.145.1
BFCL V4 multi-turn60.555.645.5
Price ($/1M in · out)$0.40 · $1.50Varies by provider$2.00 · $8.00

Comparison source ↗

This model's scores

  1. Average quality (composite, low effort)70.5
  2. τ³-bench Telecom (low effort)77.2%
  3. τ³-bench Airline (low effort)78%
  4. τ³-bench Retail (low effort)69.3%
  5. IFBench (low effort)67.4%
  6. BFCL V4 multi-turn (low effort)60.5%

Scores on a 0–100 scale (25-point gridlines); higher is better. Each benchmark links to its published source.

Pricing

Input$0.40 / 1M tokens
Cached input$0.04 / 1M tokens
Output$1.50 / 1M tokens

Inception launched Mercury Voice at 50% off: $0.20 input, $0.02 cached input and $0.75 output per million tokens.

Pricing source ↗

Strengths

  • 320 ms median and 750 ms p95 time to first answer token on OpenCall customer-service prompts at low reasoning effort, per Inception
  • Three reasoning-effort settings — low, medium and high — so a voice agent can trade latency for depth per deployment
  • Tool calling and structured outputs, with long system prompts, for agents that act during a call
  • 128K-token context window with up to 50,000 output tokens
  • OpenAI-compatible endpoint that drops into LiveKit, Pipecat, Vapi, Retell or a custom voice stack
  • Published price of $0.40 input / $1.50 output per million tokens, which Inception puts at about $0.009 per minute of conversation

Best for

  • Customer-service phone agents, where Inception's TTFAT measurements were taken on real OpenCall prompts
VideoHow OpenCall powers its voice agents with MercuryInception ↗
  • Automated drive-thru ordering — Audivi AI uses it for real-time orders, order modifications and upsells
  • Financial-services calls such as negotiating payment plans and handling objections, as Altur does
  • Appointment scheduling and other voice workflows that need tool calls without an audible pause

How to access

VideoA Mercury Voice agent booking a dental appointment in a live conversationInception ↗

FAQ

Does Mercury Voice take audio input?

No. Inception's docs list it as a text model on the v1/chat/completions endpoint. It fills the LLM slot of a voice pipeline — between speech-to-text and text-to-speech — in platforms such as LiveKit, Pipecat, Vapi and Retell.

How fast is Mercury Voice?

Inception reports a 320 ms median (p50) and 750 ms p95 time to first answer token on OpenCall customer-service prompts at low reasoning effort; at medium effort the figures are 550 ms and 1.96 s, and at high effort 1.11 s and 4.84 s. OpenCall reported median model response latency close to 170 ms on its own production workload.

What does Mercury Voice cost?

$0.40 per million input tokens, $0.04 per million cached input tokens and $1.50 per million output tokens. At launch Inception discounted it 50%, to $0.20, $0.02 and $0.75. Inception estimates about $0.009 per minute of conversation for a typical voice-agent profile.

How do I get access to Mercury Voice?

It is available to enterprise customers through the OpenAI-compatible Inception API. Inception's docs say to contact sales@inceptionlabs.ai to enable it for an organization.

What benchmarks did Inception publish for Mercury Voice?

At low effort: 77.2 on τ³-bench Telecom, 78.0 on τ³-bench Airline, 69.3 on τ³-bench Retail, 67.4 on IFBench and 60.5 on BFCL V4 multi-turn, for an average quality score of 70.5. Inception compares these with Gemma 4 31B (non-reasoning) at 59.0 and GPT-4.1 at 51.4 on the same average.