InceptionMercury Voice
Figures published by Inception, not measured independently. No benchmark publisher lists this model yet, so its price, context and latency come from its own announcement. BenchLeader runs no evaluations: nothing here has been checked by anyone else, and these numbers will be replaced the moment a publisher measures the model.
Mercury Voice is an Inception proprietary model, released 1 Oct 2026. It is not yet ranked: no independent benchmark results so far, only pricing and metadata. At $0.675 per million tokens blended it is mid-priced.
- Blended price
- $0.675/M
- $0.400 in · $1.50 out
- Output speed
- –
- First answer
- 0.32 s
- Context
- 128k
- Full answer
- –
- Cost per run
- –
- Released
- 1 Oct 2026
Results published by Inception
Inception ran these itself and published them in its launch post, at low reasoning effort. Nobody else has reproduced them. They are kept off every leaderboard and out of the index: the post reports benchmarks in versions and subsets our boards do not carry, and a maker’s own harness cannot sit beside independently measured numbers without making those boards mean less.
| Benchmark | Score | As reported |
|---|---|---|
| Average quality (composite) | 70.5 | first of the 18 configurations the maker charted, 0.6 ahead of GLM-5.3 Flash at low effort and 0.8 ahead of Gemma 4 31B reasoning |
| tau-3-bench Airline | 78.0 | — |
| tau-3-bench Telecom | 77.2 | — |
| tau-3-bench Retail | 69.3 | — |
| IFBench | 67.4 | — |
| BFCL v4 (multi-turn) | 60.5 | — |
No independent quality benchmark results yet — only pricing and metadata.
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0006 | – | – |
| Summarise a 30-page report | 12,000 / 600 | $0.0057 | – | – |
| Code edit | 6,000 / 1,500 | $0.0046 | – | – |
| Agentic coding session | 60,000 / 4,000 | $0.030 | – | – |
| Structured extraction | 2,000 / 200 | $0.0011 | – | – |
See also
Data as of 11 Oct 2026. Compare with another model.
Cite as: BenchLeader, “Mercury Voice: benchmarks, pricing, speed and rank”, https://www.benchleader.com/models/mercury-voice, data as of 11 Oct 2026.