Inference
EN
Anthropic

Claude 4 Opus Thinking

Claude 4 Opus Thinking is a reasoning model from Anthropic in the Claude 4 Opus family. 10 benchmarks count toward its index, in 6 categories.

  • Closed weights
  • Reasoning
  • Input: text
  • Released 2025-05
RankedEnough results for an index.
IndexAll categories, out of 100.47.0±7.6Median 51.0
CoverageIndex weight that has results.75%
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.1 counted51.5—
CodingCode writing and repair.None counted——
ReasoningLogic problems and puzzles.1 counted54.8—
MultimodalTasks with images and text.1 counted44.9—
KnowledgeFacts and expert knowledge.4 counted47.1155 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted39.2—
MathMath problems.2 counted43.9110 / 168

Results

10 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
MMMLUKnowledge88.8%4 / 15
Artificial Analysis MMLU-ProKnowledge87.3%5 / 15
Massive Multi-discipline Multimodal UnderstandingMultimodal76.5%12 / 28
Graduate-Level Google-Proof Q&AKnowledge79.6%68 / 141
Artificial Analysis IFBenchInstruction53.7%73 / 146
Artificial Analysis GPQA DiamondKnowledge79.6%101 / 199
Artificial Analysis Long Context ReasoningReasoning69.3%115 / 217
τ²-Bench Tool-Agent-User EvaluationAgentic73.4%79 / 149
Artificial Analysis Humanity's Last ExamKnowledge12.3%134 / 222
American Invitational Mathematics Examination 2025Math75.5%33 / 53
Artificial Analysis MATH-500Math98.2%6 / 9

Sources

BenchLMCC BY-NC 4.0

History

5 releases
Index47.1 → 47.0
23 Oct 202612 Nov 2026

Questions

How good is Claude 4 Opus Thinking?
Its index is 47.0 out of 100. That is place 168 of 283 ranked models.
What is Claude 4 Opus Thinking best at?
Knowledge: 47.1 out of 100, place 155 of 268.
What is the context window of Claude 4 Opus Thinking?
200K tokens.
Is Claude 4 Opus Thinking open source?
No. Its weights are not published.
When was Claude 4 Opus Thinking released?
May 2025.

Badge

Claude 4 Opus Thinking on Inference360: place 168 of 283, index 47.0
HTML
<a href="https://inference360.ai/models/claude-4-opus-thinking"><img src="https://inference360.ai/badges/claude-4-opus-thinking.svg" alt="Claude 4 Opus Thinking on Inference360: place 168 of 283, index 47.0" height="20"></a>
Markdown
[![Claude 4 Opus Thinking on Inference360: place 168 of 283, index 47.0](https://inference360.ai/badges/claude-4-opus-thinking.svg)](https://inference360.ai/models/claude-4-opus-thinking)

Compare with

Claude 4.1 Opus Thinking47.0MiMo-V2-Omni47.2Grok 4.1 Fast (Reasoning)47.3Claude 4 Sonnet39.5Claude 3.7 Sonnet37.7

More from Anthropic

All models
Claude 4 Sonnet39.5Claude 4 OpusUnrankedClaude Sonnet 443.0Claude 4.1 Opus Thinking47.0Claude 4.1 OpusUnranked