Compare / Claude Opus 5 vs Claude Opus 5.5

Claude Opus 5 vs Claude Opus 5.5

Which is cheaper, which is more capable, and which does better at each kind of work, from 19 results on 5 sources that measured both models.

AnthropicClaude Opus 5 Released 24 Jul 2026 · $5.00 in · $25.00 out
AnthropicClaude Opus 5.5 Released 22 Sep 2026 · $4.00 in · $20.00 out
Claude Opus 5 wins 05 too close to callClaude Opus 5.5 wins 14

Of the 19 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 or Claude Opus 5.5? The short answer

Claude Opus 5.5 is the newer version and costs less, and it scores higher (Intelligence Index 57.6 against 50.8): upgrade.

  • Claude Opus 5.5 is cheaper: $8.00 against $10.00 per million tokens, 20% less.
  • Claude Opus 5.5 scores higher on the Artificial Analysis Intelligence Index: 57.6 against 50.8.
  • Claude Opus 5.5 is newer: released 22 Sep 2026, against 24 Jul 2026 for Claude Opus 5.
  • On third-party benchmarks, on the results both models have: Claude Opus 5.5 does better for agents and tool use, professional work, reasoning and knowledge, writing and sticking to the facts.

Head to head

50.8
Intelligence Index
57.6
$10.00
Price per million tokens, blendedlower is better
$8.00
1,000,000
Context window, tokens
1,000,000

Every fact side by side

Claude Opus 5Claude Opus 5.5
Price per million tokens$5.00 in · $25.00 out$4.00 in · $20.00 out
Blended (3 in : 1 out)$10.00$8.00
Cheapest host, blended$10.00 (Amazon Bedrock)$8.00 (Amazon Bedrock)
Cached input, per million$0.50 cached input$0.20 cached input
Intelligence Index50.857.6
Spring Prompt overall–82
Output speed, tokens a second–96
Context window1,000,000 tokens1,000,000 tokens
Released24 Jul 202622 Sep 2026 (newer)
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 not measured 13 of 32
Decks from an analysisTurning a finished analysis into a deck you could present as it is not measured 7 of 32
User surveysPlanning a user survey and reading its results without being misled not measured 4 of 23
Marketing planningPlanning a year of ad spend without overspending not measured 9 of 32
Agents and tool useMulti-step tasks with tools: support desks, coding agents, function calls 9 of 159 3 of 159
Professional workReal tasks from banking, consulting and law, business documents and freelance projects 7 of 51 3 of 51
Reasoning and knowledgeHard questions across science, maths and general knowledge 7 of 187 1 of 187
WritingWhat people prefer in blind comparisons, and judged writing quality 11 of 169 5 of 169
Sticking to the factsSummarising without inventing things, and factual answers 15 of 139 3 of 139
Speed under pressureGood decisions against a real clock (fast chess) not measured 36 of 36

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 5Claude Opus 5.5
APEX-Agents reported by APEX-Agents · 1 result
Tasks passed% of tasks, higher is better 5 · 2 of 37 measured 65.8%max reasoning 73.5%max reasoning
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 3 results
Overall · TextArena rating, higher is better 11 · 4 of 387 measured 1,490high reasoning 1,504high reasoning
Overall · Text factualityArena rating, higher is better 5 · 2 of 163 measured 1,489max reasoning 1,507high reasoning
Business, management and finance · TextArena rating, higher is better 13 · 10 of 380 measured 1,486high reasoning 1,489high reasoning
Artificial Analysis reported by Artificial Analysis · 3 results
Artificial Analysis Intelligence Indexindex score, higher is better 8 · 1 of 249 measured 50.8max reasoning 57.6max reasoning
Humanity's Last Exam% of questions, higher is better 6 · 1 of 250 measured 54.9%max reasoning 61.4%max reasoning
Terminal-Bench 4.0% of tasks, higher is better 10 · 3 of 120 measured 49.0%max reasoning 59.6%max reasoning
GDP.pdf reported by GDP.pdf · 1 result
Professional document tasks% of rubric, higher is better 9 · 2 of 32 measured 24.0%max reasoning 30.6%max reasoning
SimpleQA Verified (Epoch AI) reported by SimpleQA Verified (Epoch AI) · 1 result
Correct answers% of questions, higher is better 16 · 4 of 76 measured 59.9%max reasoning 72.2%max reasoning
Show all 19 results (10 more)
MeasureClaude Opus 5Claude Opus 5.5
Arena (formerly LMArena) reported by Arena (formerly LMArena) · 8 results
Confirmed task success · AgentIPS effect estimate, higher is better 7 · 4 of 48 measured 0.10max reasoning, arena agent 0.14high reasoning, arena agent
Creative writing · TextArena rating, higher is better 11 · 2 of 385 measured 1,471high reasoning 1,516high reasoning
Expert prompts · TextArena rating, higher is better 5 · 6 of 338 measured 1,543high reasoning 1,541high reasoning
Instruction following · TextArena rating, higher is better 7 · 2 of 387 measured 1,496high reasoning 1,517high reasoning
Praise over complaint · AgentIPS effect estimate, higher is better 10 · 3 of 48 measured 0.17high reasoning, arena agent 0.31high reasoning, arena agent
Steerability · AgentIPS effect estimate, higher is better 9 · 4 of 48 measured 0.07high reasoning, arena agent 0.10high reasoning, arena agent
Tool grounding · AgentIPS effect estimate, higher is better 5 · 6 of 48 measured 0.00high reasoning, arena agent 0.00high reasoning, arena agent
Writing, literature and language · TextArena rating, higher is better 12 · 2 of 386 measured 1,480high reasoning 1,519high reasoning
Artificial Analysis reported by Artificial Analysis · 2 results
Long-context reasoning (AA-LCR)% of questions, higher is better 27 · 5 of 241 measured 82.0%medium reasoning 84.7%max reasoning
SciCode% of problems, higher is better 23 · 1 of 121 measured 56.4%max reasoning 66.9%max reasoning

Quick answers

Is Claude Opus 5 better than Claude Opus 5.5?

Claude Opus 5.5 is the newer version and costs less, and it scores higher (Intelligence Index 57.6 against 50.8): upgrade.

Which is cheaper, Claude Opus 5 or Claude Opus 5.5?

Claude Opus 5.5 is cheaper: $8.00 against $10.00 per million tokens, 20% less.

Which is better for agents and tool use, Claude Opus 5 or Claude Opus 5.5?

Claude Opus 5.5 ranks 3 of 159 models on sale for multi-step tasks with tools: support desks, coding agents, function calls, against 9 for Claude Opus 5.

Which is better for professional work, Claude Opus 5 or Claude Opus 5.5?

Claude Opus 5.5 ranks 3 of 51 models on sale for real tasks from banking, consulting and law, business documents and freelance projects, against 7 for Claude Opus 5.

Which is better for reasoning and knowledge, Claude Opus 5 or Claude Opus 5.5?

Claude Opus 5.5 ranks 1 of 187 models on sale for hard questions across science, maths and general knowledge, against 7 for Claude Opus 5.

Which is better for writing, Claude Opus 5 or Claude Opus 5.5?

Claude Opus 5.5 ranks 5 of 169 models on sale for what people prefer in blind comparisons, and judged writing quality, against 11 for Claude Opus 5.

Which is better for sticking to the facts, Claude Opus 5 or Claude Opus 5.5?

Claude Opus 5.5 ranks 3 of 139 models on sale for summarising without inventing things, and factual answers, against 15 for Claude Opus 5.

Run Claude Opus 5 and Claude Opus 5.5 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.