Inference
EN
Anthropic

Claude Opus 4.6 (Adaptive)

Claude Opus 4.6 (Adaptive) is a reasoning model from Anthropic in the Claude Opus 4.6 family. 15 benchmarks count toward its index, in 7 categories.

  • Closed weights
  • Reasoning
  • Input: text
  • Released 2026-02
RankedEnough results for an index.
IndexAll categories, out of 100.60.0±4.7Median 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.1MMedian 262K
EloLMArena rating and rank.—Median 1422

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.3 counted63.552 / 218
CodingCode writing and repair.3 counted60.071 / 209
ReasoningLogic problems and puzzles.2 counted61.062 / 227
MultimodalTasks with images and text.1 counted58.8—
KnowledgeFacts and expert knowledge.4 counted62.652 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.1 counted38.6—
MathMath problems.1 counted57.3—

Results

15 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
AIMEMath95.6% ±0.68 / 95
MMLU ProKnowledge89.1% ±0.515 / 135
Terminal-Bench 2.0 · Vals AIAgentic58.4% ±5.310 / 66
Artificial Analysis Omniscience AccuracyKnowledge47.0%30 / 164
Artificial Analysis GPQA DiamondKnowledge89.6%41 / 199
GPQA Diamond · Vals AIKnowledge89.6% ±1.228 / 135
Artificial Analysis Humanity's Last ExamKnowledge39.9%47 / 222
τ²-Bench Tool-Agent-User EvaluationAgentic92.1%35 / 149
LiveCodeBenchCoding84.7% ±1.036 / 142
Critical Physics TasksReasoning12.6%56 / 214
MMMU ProMultimodal83.9% ±0.928 / 92
Artificial Analysis Long Context ReasoningReasoning78.0%70 / 217
APEX-Agents-AAAgentic33.0%10 / 30
SWE-benchCoding78.2% ±1.830 / 87
Artificial Analysis MMMU-ProMultimodal75.4%49 / 113
Vibe Code Bench v1.1Coding53.5% ±4.763 / 123
Artificial Analysis IFBenchInstruction53.1%76 / 146
TaxEval v2Finance76.0% ±0.88 / 143
Finance Agent v1.1Finance60.0% ±2.85 / 51
CorpFin v2Finance67.0% ±0.913 / 131
MedQAHealthcare95.4% ±0.212 / 95
LegalBenchLegal85.3% ±0.422 / 149
MortgageTaxFinance68.5% ±0.915 / 97
MedScribeHealthcare86.1% ±1.924 / 109
MedCodeHealthcare49.1% ±2.126 / 106
CaseLaw v2Legal62.1% ±0.420 / 53
Artificial Analysis Global-MMLU-LiteMultilingual92.2%—

Sources

BenchLMCC BY-NC 4.0Vals AINo licence stated

History

5 releases
Index61.1 → 60.0
23 Oct 202612 Nov 2026

Questions

How good is Claude Opus 4.6 (Adaptive)?
Its index is 60.0 out of 100. That is place 74 of 283 ranked models.
What is Claude Opus 4.6 (Adaptive) best at?
Knowledge: 62.6 out of 100, place 52 of 268.
What is the context window of Claude Opus 4.6 (Adaptive)?
1M tokens.
Is Claude Opus 4.6 (Adaptive) open source?
No. Its weights are not published.
When was Claude Opus 4.6 (Adaptive) released?
February 2026.

Badge

Claude Opus 4.6 (Adaptive) on Inference360: place 74 of 283, index 60.0
HTML
<a href="https://inference360.ai/models/claude-opus-4-6-thinking"><img src="https://inference360.ai/badges/claude-opus-4-6-thinking.svg" alt="Claude Opus 4.6 (Adaptive) on Inference360: place 74 of 283, index 60.0" height="20"></a>
Markdown
[![Claude Opus 4.6 (Adaptive) on Inference360: place 74 of 283, index 60.0](https://inference360.ai/badges/claude-opus-4-6-thinking.svg)](https://inference360.ai/models/claude-opus-4-6-thinking)

Compare with

GPT-5.2-Codex59.9Muse Spark59.8Beam 1.0 501B-A23B60.2Claude Opus 4.658.0Claude Opus 4.552.0

More from Anthropic

All models
Claude Opus 4.658.0Claude Sonnet 4.657.0Claude Mythos PreviewUnrankedClaude Opus 4.5 Thinking56.3Claude Opus 4.552.0