Compare / Claude Opus 5.5 vs Mercury 2

Claude Opus 5.5 vs Mercury 2

Which is cheaper, which is more capable, and which does better at each kind of work, from 12 results on 2 sources that measured both models.

AnthropicClaude Opus 5.5 Released 22 Sep 2026 · $4.00 in · $20.00 out
InceptionMercury 2 Released 20 Feb 2026 · Not on sale
Claude Opus 5.5 wins 12Mercury 2 wins 0

Of the 12 results both models have, the better value on each. Results whose reported ranges overlap are too close to call.

Results as of 16 November 2026 (catalogue release 2026-11-16-49bdfa045dee).

Claude Opus 5.5 or Mercury 2? The short answer

Pick Claude Opus 5.5: it scores higher (Intelligence Index 57.6 against 10.8) at about the same price.

  • Claude Opus 5.5 scores higher on the Artificial Analysis Intelligence Index: 57.6 against 10.8.
  • Claude Opus 5.5 is newer: released 22 Sep 2026, against 20 Feb 2026 for Mercury 2.
  • On third-party benchmarks, on the results both models have: Claude Opus 5.5 does better for agents and tool use, reasoning and knowledge, writing and sticking to the facts.

Head to head

57.6
Intelligence Index
10.8

Every fact side by side

Claude Opus 5.5Mercury 2
Price per million tokens$4.00 in · $20.00 outNot on sale
Blended (3 in : 1 out)$8.00–
Cheapest host, blended$8.00 (Amazon Bedrock)–
Cached input, per million$0.20 cached input–
Intelligence Index57.610.8
Spring Prompt overall82–
Output speed, tokens a second96–
Context window1,000,000 tokens–
Released22 Sep 2026 (newer)20 Feb 2026
WeightsClosedClosed
Developer based inthe USthe US

Prices are the developer's list price where we have read it, otherwise the typical host on OpenRouter; see every model's price. Intelligence and speed from Artificial Analysis.

Which is better for what

Each model's rank among the models on sale today, for each kind of work, from our own benchmarks and licensed sources. Further right is better; the better rank is in green.

Product listingsTurning a sparse product feed and photos into listings that can go live 13 of 32 not measured
Decks from an analysisTurning a finished analysis into a deck you could present as it is 7 of 32 not measured
User surveysPlanning a user survey and reading its results without being misled 4 of 23 not measured
Marketing planningPlanning a year of ad spend without overspending 9 of 32 not measured
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 3 of 159 118 of 160
Professional workReal tasks from banking, consulting and law, business documents and freelance projects 3 of 51 not measured
Reasoning and knowledgeHard questions across science, maths and general knowledge 1 of 187 124 of 188
WritingWhat people prefer in blind comparisons, and judged writing quality 5 of 169 132 of 170
Sticking to the factsSummarising without inventing things, and factual answers 3 of 139 126 of 140
Speed under pressureGood decisions against a real clock (fast chess) 36 of 36 not measured

From our research

Every shared result

Each model's best result on every measure both have, grouped by benchmark, with the setting that produced it. Under each measure, every model that benchmark has measured is a grey tick, best to the right, so you can see where the two sit in the field. The headline measures of each benchmark come first; the rest are under "Show all". Green marks the better value; where a source reports ranges that overlap, neither is marked.

MeasureClaude Opus 5.5Mercury 2
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 4 · 200 of 387 measured 1,504high reasoning 1,343
Overall · Text factualityArena rating, higher is better 2 · 151 of 163 measured 1,507high reasoning 1,369
Business, management and finance · TextArena rating, higher is better 10 · 202 of 380 measured 1,489high reasoning 1,340
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 1 · 171 of 249 measured 57.6max reasoning 10.8
Humanity's Last Exam% of questions, higher is better 1 · 126 of 250 measured 61.4%max reasoning 17.1%
Terminal-Bench 4.0% of tasks, higher is better 3 · 72 of 120 measured 59.6%max reasoning 0.0%
Show all 12 results (6 more)
MeasureClaude Opus 5.5Mercury 2
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 4 results
Creative writing · TextArena rating, higher is better 2 · 221 of 385 measured 1,516high reasoning 1,292
Expert prompts · TextArena rating, higher is better 6 · 196 of 338 measured 1,541high reasoning 1,348
Instruction following · TextArena rating, higher is better 2 · 201 of 387 measured 1,517high reasoning 1,326
Writing, literature and language · TextArena rating, higher is better 2 · 218 of 386 measured 1,519high reasoning 1,308
Artificial Analysis reported by Artificial Analysis · 2 results
Long-context reasoning (AA-LCR)% of questions, higher is better 5 · 173 of 241 measured 84.7%max reasoning 43.7%
SciCode% of problems, higher is better 1 · 94 of 121 measured 66.9%max reasoning 37.7%

Quick answers

Is Claude Opus 5.5 better than Mercury 2?

Pick Claude Opus 5.5: it scores higher (Intelligence Index 57.6 against 10.8) at about the same price.

Which is better for agents and tool use, Claude Opus 5.5 or Mercury 2?

Claude Opus 5.5 ranks 3 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 118 for Mercury 2.

Which is better for reasoning and knowledge, Claude Opus 5.5 or Mercury 2?

Claude Opus 5.5 ranks 1 of 187 models on sale for hard questions across science, maths and general knowledge, against 124 for Mercury 2.

Which is better for writing, Claude Opus 5.5 or Mercury 2?

Claude Opus 5.5 ranks 5 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 132 for Mercury 2.

Which is better for sticking to the facts, Claude Opus 5.5 or Mercury 2?

Claude Opus 5.5 ranks 3 of 139 models on sale for summarising without inventing things, and factual answers, against 126 for Mercury 2.

Run Claude Opus 5.5 and Mercury 2 on your own prompt.

Benchmarks aren't your task. Test both side by side on your prompt, with every answer checked against your rules and the cost of each one.