/ Inception / Mercury Voice Inception / Mercury
Mercury Voice Text-based diffusion LLM tuned for the language stage of low-latency voice agents.
restricted Closed weights Text generation Coding
Context window 128K 128,000 tokens
Maximum output 50K 50,000 tokens
Input / 1M tokens $0.20 Inception API
Output / 1M tokens $0.75 USD · observed 15 Oct 2026
Specifications Access & pricing Caching Evaluation Sources & lifecycle
Specifications Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.
Maximum input tokens Not collected No exact model-specific contract established in the reviewed evidence.
Parameters Not collected No exact model-specific contract established in the reviewed evidence.
Knowledge cutoff Not collected No exact model-specific contract established in the reviewed evidence.
Streaming Not collected No exact model-specific contract established in the reviewed evidence.
Batch processing Not collected No exact model-specific contract established in the reviewed evidence.
Fine-tuning Not collected No exact model-specific contract established in the reviewed evidence.
Access & pricing Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.
API model ID Not collectedNo exact callable model selector established for this offering.
Access Enterprise API access by arrangement; provider documentation does not expose a confirmed exact model selector.
Regions Deployment-dependent; complete serving region inventory unverified. INPUT / 1M TOKENS $0.20
OUTPUT / 1M TOKENS $0.75
Advertised promotional snapshot on October 15, 2026. End date not published; verify current account pricing. Billed output includes reasoning. Cached input price: $0.02 / 1M tokens (base tier).
USD · observed 15 Oct 2026 · Pricing evidence (opens in a new tab) Published billing units · original currency · observed 15 Oct 2026 Category Usage Rate Unit Undiscounted reference prices Input1 $0.40 1M tokens Undiscounted reference prices Output1 $1.50 1M tokens Undiscounted reference prices Cached input1 $0.04 1M tokens
Rate conditions Crossed-out regular rate in the provider pricing table, not the advertised promotional rate. No promotion end date is published. Pricing source (opens in a new tab) Data & enterprise terms Terms permit using submissions for model training unless the user disables “Improve the model for everyone” in API Platform settings. The privacy policy uses purpose-dependent retention, not a fixed universal period. The FAQ says custom zero-retention arrangements are handled case by case; enterprise agreements may differ.
Read the service policy (opens in a new tab)
Caching Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.
01 / INITIAL INPUT Process & retain The system processes eligible input and retains reusable state.
02 / LATER REQUEST Check available state Reuse depends on matching input, session state and the cache policy.
03 / CACHE HIT Reuse & generate Reusing state can reduce repeated work. Any charges follow the billing rules below.
Support evidence (opens in a new tab) Minimum reusable input Not collected Minimum prefix length not documented.
Usage fields prompt_tokens includes cached_tokens; completion_tokens includes reasoning_tokens. Do not add nested counts again. Streaming usage requires stream_options.include_usage.Source for Usage fields (opens in a new tab) Rate conditions Advertised promotional snapshot on October 15, 2026. End date not published; verify current account pricing. Billed output includes reasoning. Crossed-out regular rate in the provider pricing table, not the advertised promotional rate. No promotion end date is published. Observed 15 Oct 2026. Write rates and storage fees are different charges; use the billing rules above.
Caching reviewed 15 Oct 2026
What the measurements say
Time to first answer token · provider voice prompts Inception · provider-reported 320ms (p50) Real customer-service prompt set with low reasoning. Reported p95 is 750ms. Not a universal SLA or time-to-first-audio guarantee.
Evaluation source (opens in a new tab) Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.
The record behind the model Why it is included Inception’s own Trust Center describes the Mercury family as proprietary. Public API and enterprise deployment access do not establish public weights.
Classification evidence (opens in a new tab) Sources reviewed Model notes Nine named current, legacy and managed records reviewed. Live discovery, endpoint docs, promotional pricing and known source conflicts retained. Earlier next-edit aliases and account-specific contracts remain open.
Enterprise access must be enabled. Mercury Voice is absent from the unauthenticated discovery list; its raw API selector is not guessed.
TTFAT includes reasoning before the first answer token. It is distinct from TTFT; the 320ms provider median and 170ms customer-workload result are not interchangeable.
Check against your workload Try representative inputs, measure task success and total cost, then check the exact endpoint’s data policy.
Read our methodology
Also from Inception Mercury 2 Mercury 2.5 Mercury Edit 2 Mercury Decide Mercury Coder Mini Mercury Coder Small Mercury 1 Mercury Router 