InceptionReleased on Feb 24, 2026

Mercury 2: Benchmarks, Pricing & Context Window

Mercury 2 is a language model from Inception, released in February 2026.

Mercury 2 is the fastest reasoning LLM, built on diffusion-based language model (dLLM) architecture. Instead of generating text token-by-token, it refines multiple text blocks simultaneously, achieving over 1,000 tokens per second on

Input
Text
Output
Text
Speed
Latency p951.66s
Throughput P951009char/s
Cost
$0.31/ 1M · 8:1 in:out
$0.25 in · $0.75 out

Mercury 2 pricing

Providers

Mercury 2 starts at $0.250 per million input tokens and $0.750 per million output tokens via Inception.

ProviderInput $/MOutput $/MMax InputMax OutputLatency p95 sThroughput P95QuantInputOutput
Inception logoInception
$0.250$0.750128.0K8.2K
1.66
1009 c/s
—

Mercury 2 API

API access coming soon

Mercury 2 will be available through our gateway shortly.

Mercury 2 examples

Recent arena outputs from Mercury 2, picked from the highest-ranked matchups.

Mercury 2 license

Mercury 2 is released under the Proprietary license, which restricts commercial use.

License
Proprietary
Non-commercial

Proprietary license - usage restrictions apply

FAQ

Common questions about Mercury 2.

When was Mercury 2 released?

Mercury 2 was released on February 24, 2026 by Inception. This is the official Mercury 2 release date tracked on LLM Stats.

Is Mercury 2 available via API?

Yes, Mercury 2 is available via API. See the official documentation for authentication and endpoint details. It is served by 1 provider tracked on LLM Stats.

Who created Mercury 2?

Mercury 2 was created by Inception.

What is the license for Mercury 2?

Mercury 2 is released under the Proprietary license.

Where can I use Mercury 2?

Mercury 2 is available through 1 provider including Inception.

Where is the Mercury 2 paper or technical report?

Mercury 2 has a paper or technical report available at https://www.inceptionlabs.ai/blog/introducing-mercury-2. Use that source for architecture, training, release and evaluation details.