Mercury 2 vs Mercury 3

Side by side: price, limits, capabilities and real cost examples. Catalog data · checked Nov 16, 2026

Short answer

  • Mercury 3 is about 2.3× cheaper than Mercury 2 for a typical mix of three parts input and one part output.
  • Mercury 3 reads 7.8× more text per request (1M vs 128K tokens).
  • A quality verdict needs a hands-on test. Neither model has completed our Hero suite yet, so this page compares facts, not output quality.
Mercury 2InceptionMercury 3Inception
Input price / 1M tokens$0.25$0.1
Output price / 1M tokens$0.75$0.35
Context window128K1M
Max output per answer50K131.1K
ReleasedFeb 2026Oct 2026
Tool callingYesYes
Structured outputYesYes
Reasoning modeYesYes
Prompt cachingYesYes
What it costs in practice
1,000 support replies1,000 chats with 800 input and 250 output tokens each$0.388$0.168
Summarize a 300-page reportabout 150,000 input tokens in one request, 2,000 tokens out$0.039$0.016
10,000 product descriptions300 input and 200 output tokens per text$2.25$1
One day of agentic codingabout 3M input and 150K output tokens across many calls$0.863$0.353

Green marks the better value in that row. Prices are first-party list prices in USD. Method.

Calculate with my numbersCompare with another model