Skip to content
ModelCap

Model decision surface

Compare AI models

Start with Mercury 2.5 and Nemotron Cascade 2 30B A3B, or choose any two current ranked language models. Compare capability evidence, price, context, provider availability, and weight access without pretending one field decides every use case.

Current public data

Mercury 2.5 vs Nemotron Cascade 2 30B A3B

Live dataset updated 9/9/2026, 8:46:14 PM UTC

Open 1200×630 evidence receipt
Factual comparison of Mercury 2.5 and Nemotron Cascade 2 30B A3B
Field
Mercury 2.5

Inception Labs

ModelCap position#111#108
Index score37.338.4
EvidenceEstimatedglobal-corpus-prior over 182 held-out anchors (23%); opens from measured predecessor inception/mercury-2 (78%); shrunk 4.7 up toward the measured corpusEstimatedpublisher-corpus-prior over 182 held-out anchors (57%); launch card against 3 resolved peers on 11 rows (43%); exceeds every named peer on 3 of 11 rows; optimism haircut 0 from cross-lab-probe; shrunk 4.9 up toward the measured corpus
Input / 1M$0.04Unavailable
Output / 1M$0.15Unavailable
Pricing statusfreshunavailable
Context260K—
Providers10
Weight accessAPI onlyRestricted license

Decision facts

  • Mercury 2.5 is #111; Nemotron Cascade 2 30B A3B is #108 on the same current language board.
  • Index scores are 37.3 for Mercury 2.5 and 38.4 for Nemotron Cascade 2 30B A3B. Their published uncertainty intervals overlap, so the rank difference alone does not establish a reliable capability advantage for your workload.
  • Both positions use Estimated evidence.
  • Weight access differs: Mercury 2.5 is none; Nemotron Cascade 2 30B A3B is restricted.
  • ModelCap currently lists 1 providers for Mercury 2.5 and 0 for Nemotron Cascade 2 30B A3B.

These are separate published fields, not a synthetic winner. ModelCap does not collapse price, access, context, and capability evidence into a hidden recommendation score.

Each comparison page is a permanent, shareable URL with the same live figures as this tool: ModelCap Index position, API pricing, context window, provider count, weight access and every shared benchmark board.