Anthropicproprietary
Released: 2026-09-21Claude Opus 5.5
Anthropic Opus 5.5 high-capacity reasoning model engineered for long-running autonomous knowledge and software refactoring.
Intelligence Index
96.2/ 100
Calibrated multi-domain compositeThroughput Speed
52tok/sec
Output streaming throughputTime To First Token
460ms
Initial chunk server latencyLiveBench Contam-Free
83.5%
September 2026 suite scoreSourced Benchmark Evaluations (61)
Verified performance across authoritative benchmarks with provenance tracking.| Benchmark | Status | Score | Trust | Date | Source Type | Provenance |
|---|---|---|---|---|---|---|
| RULER | Nearing Saturation | 76.2% | 84 | 2026-09-12 | independent | Source ↗ |
| Mercury | Deprecated | 72.1% | 10 | 2026-09-12 | independent | Source ↗ |
| MGSM | Nearing Saturation | 74.8% | 65 | 2026-09-12 | independent | Source ↗ |
| MMLU | Saturated | 84.5% | 25 | 2026-09-12 | independent | Source ↗ |
| MMLU-Pro | Nearing Saturation | 76.2% | 75 | 2026-09-12 | independent | Source ↗ |
| MMLU-Redux | Saturated | 81.1% | 25 | 2026-09-12 | independent | Source ↗ |
| MT-Bench | Nearing Saturation | 7.8% | 75 | 2026-09-12 | independent | Source ↗ |
| MultiPL-E | Deprecated | 75.3% | 10 | 2026-09-12 | independent | Source ↗ |
| Needle-in-a-Haystack | Saturated | 96.6% | 44 | 2026-09-12 | independent | Source ↗ |
| OlympiadBench | Active | 64.5% | 80 | 2026-09-12 | independent | Source ↗ |
| Omni-MATH | Deprecated | 58.2% | 10 | 2026-09-12 | independent | Source ↗ |
| OpenAI-MRCR | Active | 69.6% | 80 | 2026-09-12 | independent | Source ↗ |
| PutnamBench | Active | 40.9% | 90 | 2026-09-12 | independent | Source ↗ |
| QASPER | Nearing Saturation | 68.7% | 65 | 2026-09-12 | independent | Source ↗ |
| QuALITY | Nearing Saturation | 72.9% | 78 | 2026-09-12 | independent | Source ↗ |
| AGIEval | Saturated | 70.3% | 25 | 2026-09-12 | independent | Source ↗ |
| SCROLLS | Nearing Saturation | 66.2% | 65 | 2026-09-12 | independent | Source ↗ |
| SimpleQA | Active | 43.1% | 90 | 2026-09-12 | independent | Source ↗ |
| Soohak | Active | 59.8% | 98 | 2026-09-12 | independent | Source ↗ |
| StructFlowBench | Nearing Saturation | 64.7% | 75 | 2026-09-12 | independent | Source ↗ |
| SuperGLUE | Saturated | 83.4% | 25 | 2026-09-12 | independent | Source ↗ |
| SWE-bench | Deprecated | 55.7% | 10 | 2026-09-12 | independent | Source ↗ |
| SWE-bench Pro | Active | 50.8% | 90 | 2026-09-12 | independent | Source ↗ |
| SWE-bench Verified | Saturated | 63.9% | 25 | 2026-09-12 | independent | Source ↗ |
| SWE-Lancer | Deprecated | 55.7% | 20 | 2026-09-12 | independent | Source ↗ |
| TabMWP | Saturated | 72.4% | 25 | 2026-09-12 | independent | Source ↗ |
| Terminal-Bench | Active | 57.3% | 80 | 2026-09-12 | independent | Source ↗ |
| TruthfulQA | Saturated | 72.9% | 25 | 2026-09-12 | independent | Source ↗ |
| WinoGrande | Saturated | 85.1% | 28 | 2026-09-12 | independent | Source ↗ |
| ZeroSCROLLS | Nearing Saturation | 66.2% | 78 | 2026-09-12 | independent | Source ↗ |
| MBPP | Saturated | 86.8% | 25 | 2026-09-12 | independent | Source ↗ |
| AgentIF | Nearing Saturation | 51.8% | 80 | 2026-09-12 | independent | Source ↗ |
| AIME | Nearing Saturation | 74.8% | 75 | 2026-09-12 | independent | Source ↗ |
| ARC-AGI | Saturated | 76.9% | 25 | 2026-09-12 | independent | Source ↗ |
| ARC-AGI-2 | Nearing Saturation | 63.8% | 75 | 2026-09-12 | independent | Source ↗ |
| ARC-AGI-3 | Active | 53.2% | 93 | 2026-09-12 | independent | Source ↗ |
| BIG-Bench Hard | Saturated | 80.2% | 25 | 2026-09-12 | independent | Source ↗ |
| CodeContests | Deprecated | 53.2% | 10 | 2026-09-12 | independent | Source ↗ |
| CodeXGLUE | Deprecated | 77% | 23 | 2026-09-12 | independent | Source ↗ |
| CommonsenseQA | Saturated | 85.9% | 28 | 2026-09-12 | independent | Source ↗ |
| ComplexBench | Nearing Saturation | 62.5% | 80 | 2026-09-12 | independent | Source ↗ |
| CyberSecEval | Deprecated | 73.7% | 10 | 2026-09-12 | independent | Source ↗ |
| FireBench | Active | 64% | 95 | 2026-09-12 | independent | Source ↗ |
| FrontierCode | Active | 47.5% | 98 | 2026-09-12 | independent | Source ↗ |
| FrontierMath | Active | 22% | 98 | 2026-09-12 | independent | Source ↗ |
| GLUE | Saturated | 88.3% | 28 | 2026-09-12 | independent | Source ↗ |
| GPQA | Nearing Saturation | 72.9% | 78 | 2026-09-12 | independent | Source ↗ |
| GSM-Hard | Deprecated | 75.6% | 10 | 2026-09-12 | independent | Source ↗ |
| GSM8K | Saturated | 85% | 25 | 2026-09-12 | independent | Source ↗ |
| HumanEval | Saturated | 93.4% | 25 | 2026-09-12 | independent | Source ↗ |
| HumanEval+ | Saturated | 88.5% | 25 | 2026-09-12 | independent | Source ↗ |
| Humanity's Last Exam | Active | 36% | 93 | 2026-09-12 | independent | Source ↗ |
| IFEval | Deprecated | 74.7% | 25 | 2026-09-12 | independent | Source ↗ |
| InfiniteBench | Nearing Saturation | 70.4% | 65 | 2026-09-12 | independent | Source ↗ |
| LiveCodeBench | Saturated | 67.2% | 39 | 2026-09-12 | independent | Source ↗ |
| LongBench | Nearing Saturation | 71.2% | 65 | 2026-09-12 | independent | Source ↗ |
| LongBench-Pro | Active | 67.9% | 80 | 2026-09-12 | independent | Source ↗ |
| MATH | Saturated | 77.1% | 25 | 2026-09-12 | independent | Source ↗ |
| MathArena | Active | 63% | 99 | 2026-09-12 | independent | Source ↗ |
| MBPP+ | Nearing Saturation | 83.5% | 65 | 2026-09-12 | independent | Source ↗ |
| LiveBench (Continuous) | Active | 83.5% | 94 | 2025-02-01 | 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.