Inference
EN
Anthropic

Claude Opus 4.5 Thinking

Claude Opus 4.5 Thinking is a reasoning model from Anthropic in the Claude Opus 4.5 family. 19 benchmarks count toward its index, in 7 categories.

  • Closed weights
  • Reasoning
  • Input: text
  • Released 2025-11
RankedEnough results for an index.
IndexAll categories, out of 100.56.3±4.2Median 51.0
CoverageIndex weight that has results.95%
SpeedOutput tokens per second.—Median 67/s
First tokenSeconds to the first output token.—Median 1.3 s
Input / 1MUS dollars per 1M input tokens.—Median $0.60
Output / 1MUS dollars per 1M output tokens.—Median $2.55
ContextMaximum tokens in one request.200KMedian 262K
EloLMArena rating and rank.—Median 1422

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.4 counted59.769 / 218
CodingCode writing and repair.4 counted54.4102 / 209
ReasoningLogic problems and puzzles.4 counted54.289 / 227
MultimodalTasks with images and text.1 counted57.3—
KnowledgeFacts and expert knowledge.4 counted59.070 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted43.2—
MathMath problems.1 counted57.2—

Results

19 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
Terminal-Bench 1.0Agentic57.5% ±5.55 / 47
AIMEMath95.4% ±0.412 / 95
Artificial Analysis MMLU-ProKnowledge89.5%2 / 15
Artificial Analysis Omniscience AccuracyKnowledge46.6%32 / 164
MMLU ProKnowledge87.3% ±0.429 / 135
Terminal-Bench 2.0 · Vals AIAgentic53.9% ±5.316 / 66
τ²-Bench Tool-Agent-User EvaluationAgentic89.5%43 / 149
Artificial Analysis GPQA DiamondKnowledge86.6%63 / 199
LiveCodeBenchCoding83.7% ±1.046 / 142
GPQA Diamond · Vals AIKnowledge85.9% ±2.544 / 135
Artificial Analysis LiveCodeBenchCoding87.1%4 / 12
Artificial Analysis Long Context ReasoningReasoning77.3%77 / 217
Artificial Analysis Humanity's Last ExamKnowledge30.1%79 / 222
IOI v1Coding20.3% ±5.223 / 61
MMMU ProMultimodal82.9% ±0.936 / 92
Critical Physics TasksReasoning4.6%87 / 214
SWE-benchCoding76.4% ±1.937 / 87
Artificial Analysis IFBenchInstruction58.0%65 / 146
ARC-AGI-2 (semi-private)Reasoning22.8%59 / 131
ARC-AGI-1 (semi-private)Reasoning73.9%58 / 126
Terminal-Bench HardAgentic47.0%12 / 23
Artificial Analysis MMMU-ProMultimodal74.0%61 / 113
Vibe Code Bench v1.1Coding20.6% ±3.289 / 123
MGSMMultilingual95.2% ±1.31 / 74
MedQAHealthcare95.9% ±0.210 / 95
Finance Agent v1.1Finance58.8% ±2.88 / 51
TaxEval v2Finance74.9% ±0.826 / 143
LegalBenchLegal84.6% ±0.431 / 149
MortgageTaxFinance67.7% ±0.922 / 97
MedCodeHealthcare49.2% ±2.025 / 106
CorpFin v2Finance65.1% ±0.932 / 131
MedScribeHealthcare85.3% ±1.929 / 109
CaseLaw v2Legal62.6% ±0.118 / 53
ARC-AGI-2 (public eval)Reasoning24.2%51 / 122
ARC-AGI-1 (public eval)Reasoning81.6%50 / 116
Artificial Analysis Global-MMLU-LiteMultilingual91.3%—
Artificial Analysis AIME 2025Math91.3%—

Sources

BenchLMCC BY-NC 4.0Vals AINo licence statedARC Prize FoundationTerms restrict reuse

History

5 releases
Index57.3 → 56.3
23 Oct 202612 Nov 2026

Questions

How good is Claude Opus 4.5 Thinking?
Its index is 56.3 out of 100. That is place 93 of 283 ranked models.
What is Claude Opus 4.5 Thinking best at?
Knowledge: 59.0 out of 100, place 70 of 268.
What is the context window of Claude Opus 4.5 Thinking?
200K tokens.
Is Claude Opus 4.5 Thinking open source?
No. Its weights are not published.
When was Claude Opus 4.5 Thinking released?
November 2025.

Badge

Claude Opus 4.5 Thinking on Inference360: place 93 of 283, index 56.3
HTML
<a href="https://inference360.ai/models/claude-opus-4-5-thinking"><img src="https://inference360.ai/badges/claude-opus-4-5-thinking.svg" alt="Claude Opus 4.5 Thinking on Inference360: place 93 of 283, index 56.3" height="20"></a>
Markdown
[![Claude Opus 4.5 Thinking on Inference360: place 93 of 283, index 56.3](https://inference360.ai/badges/claude-opus-4-5-thinking.svg)](https://inference360.ai/models/claude-opus-4-5-thinking)

Compare with

MiMo-V2.5-Pro56.3GLM-5.156.0MiniMax M355.7Claude Sonnet 4.657.0Claude Opus 4.552.0

More from Anthropic

All models
Claude Opus 4.552.0Claude Haiku 4.5 Thinking43.5Claude Haiku 4.543.2Claude Sonnet 4.5 Thinking53.8Claude Sonnet 4.547.9