Skip to main content
AI Choice Engine

Inception

Inception Mercury 2

Inception Mercury 2 is a Inception model with 128K context, $0.25 in / $0.75 out per million tokens, last verified October 25, 2026. Open weights: not verified. Status: Superseded; the current pick is Inception Mercury 2.5. Public-board scores are linked to the publisher; vendor-only figures stay in the watch-out, not on the board.

Mercury 2 is Inception's diffusion language model, with a published text API and vendor-reported Blackwell throughput.

Recorded profile verification October 25, 2026Individual provider fields may change

AI model specification details
SpecificationInception Mercury 2
ProviderInception
TierBalanced
Context window128K
Max output50K
Input / 1M tokens$0.25
Output / 1M tokens$0.75
WeightsNot verifiedUnverified
ParametersNot verifiedUnverified
Reasoning levelsinstant, low, medium, high
Modalitiestext
API model idmercury-2
ReleasedNot verifiedUnverified

Pricing tiers: $0.25/$0.75 per MTok (cached input $0.025); closed API. 128K context, 50,000 max output. Diffusion-based model; the vendor reports 1,009 tokens/s on NVIDIA Blackwell, not a guaranteed service rate.

Best for

  • Ultra-low-latency
  • High-throughput generation
  • Structured output

Watch out

Current docs call Mercury 2 previous generation. Instant/low/medium/high effort is explicitly supported, with medium default. Exact release date remains unverified: September 29 on the announcement extraction belongs to the Mercury Voice related card, not this model. Blackwell throughput depends on workload. A published hosted API does not establish public redistribution weights/licence.

Source receipts

Catalog figures for this model were checked against the following sources.

Common questions

Inception Mercury 2

Answered from the verified figures on this page rather than general guidance.

How much does Inception Mercury 2 cost per million tokens?
Inception Mercury 2 lists base rates of $0.25 per million input tokens and $0.75 per million output tokens. Estimate both input and output token volume and any additional charges for your workload. $0.25/$0.75 per MTok (cached input $0.025); closed API. 128K context, 50,000 max output. Diffusion-based model; the vendor reports 1,009 tokens/s on NVIDIA Blackwell, not a guaranteed service rate.
What is Inception Mercury 2's context window?
Inception Mercury 2 accepts about 128K tokens of context. That only matters if you routinely send very long documents, large codebases, or multi-turn histories that approach that limit.
What is Inception Mercury 2 best for?
Inception Mercury 2 is a balanced tier from Inception. It suits ultra-low-latency, high-throughput generation, structured output. Current docs call Mercury 2 previous generation. Instant/low/medium/high effort is explicitly supported, with medium default. Exact release date remains unverified: September 29 on the announcement extraction belongs to the Mercury Voice related card, not this model. Blackwell throughput depends on workload. A published hosted API does not establish public redistribution weights/licence.
Model comparison catalogBenchmark library