Anthropicproprietary
Released: 2026-09-22Artificial AnalysisClaude Opus 5.5
Anthropic Opus 5.5 high-capacity reasoning model engineered for long-running autonomous knowledge and software refactoring.
Intelligence Index
59.1/ 100
Artificial Analysis Intelligence Index (independent composite)Throughput Speed
114tok/sec
Output streaming throughputTime to First Answer
405s
Time to first answer token, including any thinkingCost to Run the Index
10.04USD
Artificial Analysis cost of the full Index runMeasured results · Artificial Analysis
Independent evaluations of Claude Opus 5.5 (Max, Default Fallback configuration). Values are reported as published; nothing here is estimated or derived. Snapshot 2026-11-15.
- Intelligence Index
- 59.1
- Cost to run the Index
- $10.04
- Time per task
- 25.4 min
- Omniscience index
- 48.1
- Reasoning model
- Yes
- Cached-input price
- $0.2 / 1M
Reasoning & knowledge
- AA-Omniscience Accuracy65.5%
- CritPt31.7%
- CritPt v2 (Hard)68.0%
- HLE-Diamond67.9%
- Humanity's Last Exam61.4%
Coding & terminal agents
- FrontierCode 1.152.2%
- FrontierCode-AA 2.046.4%
- SciCode66.9%
- Terminal-Bench 4.059.6%
- Terminal-Bench-Science 0.159.1%
Agentic work
- AA-AnalystAgent56.3%
- AA-Briefcase v1.1 (Elo)1801 Elo
- AutomationBench-AA69.5%
- GDPval-AA (normalized Elo)1866 Elo
- Harvey LAB-AA4.2%
- ITBench-AA38.2%
- Multimodal Knowledge Agent (MKA)53.0%
Long context & documents
Health & safety
Reasoning-effort variants evaluated
| Variant | Intelligence Index | Cost to run Index | Output speed (tok/s) |
|---|---|---|---|
| Max, Default Fallback | 59.1 | $10.04 | 113.5 |
| Xhigh, Default Fallback | 57.4 | $6.05 | 109.3 |
| High, Default Fallback | 51.8 | $3.46 | 107.0 |
| Medium, Default Fallback | 48.2 | $2.45 | 105.4 |
| Low, Default Fallback | 34.8 | $0.98 | 103.8 |
Headline Evaluation Breakdown
Six independently measured evaluations (Artificial Analysis)Reasoning · CritPt v2 (Hard)68.0%
Coding & agents · Terminal-Bench 4.059.6%
Long context · AA-LCR84.7%
Knowledge · Humanity's Last Exam61.4%
Sourced Benchmark Evaluations (23)
Verified performance across authoritative benchmarks with provenance tracking.| Benchmark | Status | Score | Trust | Date | Source Type | Provenance |
|---|---|---|---|---|---|---|
| AA-Omniscience (Non-Hallucination) | Active | 49.7% | 93 | 2026-11-15 | independent | Source ↗ |
| VIALS-AA | Active | 61.5% | 93 | 2026-11-15 | independent | Source ↗ |
| ITBench-AA | Active | 38.2% | 93 | 2026-11-15 | independent | Source ↗ |
| Harvey LAB-AA | Active | 4.2% | 93 | 2026-11-15 | independent | Source ↗ |
| AA-AnalystAgent | Active | 56.3% | 93 | 2026-11-15 | independent | Source ↗ |
| AutomationBench-AA | Active | 69.5% | 93 | 2026-11-15 | independent | Source ↗ |
| Multimodal Knowledge Agent (MKA) | Active | 53% | 93 | 2026-11-15 | independent | Source ↗ |
| MLCR-AA (Medical Long Context Reasoning) | Active | 27.8% | 93 | 2026-11-15 | independent | Source ↗ |
| GDP.pdf | Active | 26.2% | 93 | 2026-11-15 | independent | Source ↗ |
| FrontierCode-AA 2.0 | Active | 46.4% | 93 | 2026-11-15 | independent | Source ↗ |
| HealthBench Professional | Active | 59.9% | 93 | 2026-11-15 | independent | Source ↗ |
| HealthBench | Active | 57% | 93 | 2026-11-15 | independent | Source ↗ |
| FrontierCode | Active | 52.2% | 98 | 2026-11-15 | independent | Source ↗ |
| AA-Omniscience (Accuracy) | Active | 65.5% | 93 | 2026-11-15 | independent | Source ↗ |
| HLE-Diamond | Active | 67.9% | 93 | 2026-11-15 | independent | Source ↗ |
| CritPt v2 (Hard) | Active | 68% | 93 | 2026-11-15 | independent | Source ↗ |
| CritPt | Active | 31.7% | 93 | 2026-11-15 | independent | Source ↗ |
| AA-LCR (Long Context Reasoning) | Active | 84.7% | 93 | 2026-11-15 | independent | Source ↗ |
| Terminal-Bench-Science 0.1 | Active | 59% | 93 | 2026-11-15 | independent | Source ↗ |
| Terminal-Bench 4.0 | Active | 59.6% | 93 | 2026-11-15 | independent | Source ↗ |
| SciCode | Active | 66.9% | 93 | 2026-11-15 | independent | Source ↗ |
| MMMU-Pro | Active | 87.7% | 93 | 2026-11-15 | independent | Source ↗ |
| Humanity's Last Exam | Active | 61.4% | 93 | 2026-11-15 | independent | Source ↗ |
Anthropic Python SDKAPI Ready
import anthropic
client = anthropic.Anthropic()
message = client.messages.create(
model="claude-opus-5-5",
max_tokens=2048,
temperature=0,
messages=[{"role": "user", "content": "Execute SWE-bench verified repository fix"}]
)
print(message.content[0].text)Evaluation & Sourcing Notes
Scores listed for Claude Opus 5.5 represent verified evaluations extracted from official research papers, independent evaluation suites (HELM, LMSYS, OpenCompass, LiveBench), and verified audit reports.
All benchmarks marked as saturated or deprecated reflect historical performance where the benchmark no longer provides active discriminative power.