Skip to main content
Radar

Five or fewer confirmed AI changes, with original sources, on mornings when something changed.A free source-linked morning brief.

See the free Radar Brief
BenchLM

Mercury 2.5 Preview

CurrentReleased Aug 31, 2026ProprietaryReasoning260K context

Released Aug 31, 2026 — see all recent releases

Decision reading
Mercury 2.5 Preview is tracked, but not publicly ranked yet. The profile exposes 0 sourced benchmark rows and leaves unsupported fields blank until a published record exists.

Data as of September 1, 2026 · How the score is built

Compare Mercury 2.5 PreviewFind alternatives

Spec sheet

Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.

API model ID
inception/mercury-2.5-previewInception model and pricing documentation
Context window
260KInception model and pricing documentation
Maximum output
Not sourced yet
Knowledge cutoff
Not sourced yet
Input modalities
Not sourced yet
Output modalities
Not sourced yet
Parameters
Not disclosed by the provider
Availability
Available through OpenRouter as `inception/mercury-2.5-preview` with text input and output, optional reasoning levels, parallel tool calling, structured JSON output, a 260,000-token context window, and up to 65,536 completion tokens.
Cloud regions
Not tracked yet
Lifecycle
Current
API capabilities
Tool calling, structured outputs, and batch support are not tracked yet
Prompt caching
Not documented in the pricing recordInception model and pricing documentation
Self-host
Weights are not published
Rate limits
Not tracked yet

Lineage

The sequence follows explicit supersedes links. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.

  1. Feb 24, 2026

    Mercury 2

    Score 48.1 · $0.25 / $0.75

  2. Aug 31, 2026 · you are here

    Mercury 2.5 Preview

    Not publicly ranked · Price not listed

preview · Preview

How to read this profile

The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.

We track Mercury 2.5 Preview, but no weighted text-model benchmark result is published on the site yet. This page shows the metadata and separate protocol evidence we can verify now; a BenchLM score will appear only if compatible public evaluations land.

Mercury 2.5 Preview is a proprietary model with a 260K context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.

Available through OpenRouter as `inception/mercury-2.5-preview` with text input and output, optional reasoning levels, parallel tool calling, structured JSON output, a 260,000-token context window, and up to 65,536 completion tokens.

OpenRouter's August 31, 2026 launch listing describes Mercury 2.5 Preview as Inception's next diffusion reasoning model and reports 1,107 output tokens per second plus a 10-point intelligence improvement over Mercury 2. BenchLM does not map either marketing claim into weighted benchmark fields because the listing provides no exact, benchmark-native result table. The model remains outside the rankings until source-backed benchmark coverage is available.

Its explicit predecessor is Mercury 2. The profile has no source-displayable benchmark row yet.

Radar

Mercury 2.5 Preview release history

Full release history

Radar confirmed these at the source. Already tracking Mercury 2.5 Preview? See the free Radar Brief for what changes next.

Frequently asked questions

How does Mercury 2.5 Preview perform overall in AI benchmarks?

Mercury 2.5 Preview does not have any source-displayable benchmark rows yet, so this profile does not assign a public score or rank. Documented specifications remain visible, while score-led charts and claims stay unavailable until a published evaluation can be attached to the exact model.

Does Mercury 2.5 Preview have full benchmark coverage on BenchLM?

No. Mercury 2.5 Preview currently has 15 source-displayable rows across 408 tracked benchmark slots. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.

What is the context window size of Mercury 2.5 Preview?

Mercury 2.5 Preview has a documented context window of 260K. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.

Compare Mercury 2.5 Preview with every tracked model405 comparisons

Last updated September 1, 2026. Runtime fields remain blank until a sourced snapshot exists.

Watch Mercury 2.5 Preview in the weekly brief

Get one weekly email when material rank, price, availability, or benchmark evidence changes are worth revisiting.

Read a sample issue

Join 2,000+ readers.