Humanity's Last ExamVerified
61.4%
HLE (AA run)
Models / Anthropic
Claude Opus 5.5
Claude Opus · Released 2026-09-22
Anthropic's 22 Sep 2026 Opus. API list $4 / $20 per 1M tokens, 1M context, 128K output. Text and image in, text out. No AA or Vals ledger row on this page as of 23 Sep 2026.
Current Claude Opus line for long-running coding and knowledge work on the Claude API, Claude apps, and major clouds.
Composite70
Context1M
Input / 1M$4.0
Output / 1M$20
Knowledge cutoffJun 2026
Output speed96 t/s
Statusflagship
Family profile
Best published score in each covered benchmark family.
71/100 avg
- Reasoning
- 61
- Coding
- 67
- Long context
- 85
3 tested benchmark families
Editor's note
Recorded from the 22 Sep 2026 Anthropic launch page and the Claude Platform overview (1M context, 128K output, June 2026 cutoff). Re-checked 28 Sep 2026: list, context, and availability unchanged. Vendor benchmark claims stay off the ledger. Haiku 5.5 is not a published model id. Editorial depth still needs a desk pass.Benchmark placements
Per-benchmark positions
Where Claude Opus 5.5 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
| Benchmark | Family | Rank | Score | Source | Date |
|---|---|---|---|---|---|
| Humanity's Last Exam | Reasoning | #1/ 186 | 61 | Artificial Analysis | 2026-09-28 |
| SciCode | Coding | #1/ 97 | 67 | Artificial Analysis | 2026-09-28 |
| AA-LCR | Long context | #4/ 184 | 85 | Artificial Analysis | 2026-09-28 |
| AA output speed | Performance | #60/ 195 | Artificial Analysis | 2026-10-12 |
Family context
Strongest rows in each family
The three highest-scoring models with pages in each capability family. Where Claude Opus 5.5 shows up, it's highlighted.
ReasoningClaude Opus 5.5 · 61
- 196
- 295
- 395
Math
- 152
- 252
- 350
CodingClaude Opus 5.5 · 67
- 174
- 273
- 370
Long contextClaude Opus 5.5 · 85
- 189
- 286
- 385
Newest receipts
4 recentLast 120 days
Benchmark rows added to the public ledger for Claude Opus 5.5 in the last 120 days. Older rows live in the full table below.
| Benchmark | Family | Score | Source | Days ago |
|---|---|---|---|---|
| AA output speed | Performance | 96 t/s | Artificial Analysis | 4 |
| AA-LCR | Long context | 84.67% | Artificial Analysis | 18 |
| Humanity's Last Exam | Reasoning | 61.4% | Artificial Analysis | 18 |
| SciCode | Coding | 66.9% | Artificial Analysis | 18 |
Family coverage
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Price vs. performance
Where Claude Opus 5.5 sits on the scatter
Hover any dot for name, score, and input price. Keyboard: tab through the top twelve, or use the ranking below.
AnthropicOpenAIGoogleAlibabaMetaZhipuxAIDeepSeekMiniMaxXiaomiMoonshotOtherMeituanMistralNVIDIACohere
Full benchmark ledger
Every catalog benchmark for Claude Opus 5.5. Scores link to the original source; gaps mean no public row exists yet.
4 of 13 catalog benchmarks have a sourced row for Claude Opus 5.5.
Reasoning
Coding
SciCodeVerified
66.9%
SciCode (AA run)
Long context
AA-LCRVerified
84.67%
AA-LCR (AA run)
Performance
AA output speedVerified
96 t/s
Median output tokens/s (1k prompt, default provider)
Changelog
Version history
- Still the current Opus line on the Claude API and major clouds.
- Public on the Claude API, Claude apps, and AWS, Google Cloud, and Microsoft Azure.
Sources
Where the numbers come from
- Anthropic Claude Opus 5.5 announcement· checked 2026-09-28
- Claude Platform Opus 5.5 overview· checked 2026-09-28
- Claude Platform pricing· checked 2026-09-28
Questions this model page answers
Every answer below is assembled from the dated fields on this page. Nothing is written separately for search.
What is Claude Opus 5.5?
Claude Opus 5.5 is a Anthropic model: Anthropic's 22 Sep 2026 Opus. API list $4 / $20 per 1M tokens, 1M context, 128K output. Text and image in, text out. No AA or Vals ledger row on this page as of 23 Sep 2026.
What is Claude Opus 5.5 for?
Current Claude Opus line for long-running coding and knowledge work on the Claude API, Claude apps, and major clouds.
What does Claude Opus 5.5 cost?
Claude Opus 5.5 lists $4.0 input and $20 output per 1M tokens. Pricing last checked 2026-09-28.
How large is the Claude Opus 5.5 context window?
Claude Opus 5.5 publishes a 1M token context window.
What is the VerdictPal status of Claude Opus 5.5?
Claude Opus 5.5 carries Flagship status on the model atlas.
The Pack · Editorial newsletter
Know when the rules move.
One short email when a university changes its AI rule for theses, or a research tool changes its pricing or privacy terms. No tracking, no third-party analytics. Unsubscribe in one click.