Inference
EN
Anthropic

Claude 3 Opus

Claude 3 Opus is a standard model from Anthropic. 8 benchmarks count toward its index, in 4 categories.

  • Closed weights
  • Standard
  • Input: text
  • Released 2024-03
RankedEnough results for an index.
IndexAll categories, out of 100.23.3±9.7Median 51.0
CoverageIndex weight that has results.60%
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.$15.00Median $0.60
Output / 1MUS dollars per 1M output tokens.$75.00Median $2.55
ContextMaximum tokens in one request.200KMedian 262K
EloLMArena rating and rank.1262±3 #294Median 1422

Capabilities

Index per category, out of 100.
Out of 100
AgenticMulti-step tasks with tools.None counted——
CodingCode writing and repair.2 counted9.5209 / 209
ReasoningLogic problems and puzzles.1 counted37.0—
MultimodalTasks with images and text.None counted——
KnowledgeFacts and expert knowledge.3 counted27.6247 / 268
MultilingualTasks in many languages.None counted——
InstructionTasks with strict prompt rules.None counted——
MathMath problems.2 counted14.4165 / 168

Results

8 counted
BenchmarkThe benchmark name.ResultThe score from the source.PlacePlace among results on this benchmark.
MATH level 5Math37.5% ±1.161 / 88
GPQA diamond · Epoch AIKnowledge47.2% ±2.6156 / 217
SWE-bench Lite · swebench.comCoding8.0%11 / 15
OTIS Mock AIME 2024-2025Math4.7% ±2.0171 / 200
SWE-bench Lite · SWE-bench/experimentsCoding4.3%13 / 15
Artificial Analysis GPQA DiamondKnowledge48.9%177 / 199
SWE-bench Verified · swebench.comCoding11.4%41 / 46
SimpleQA VerifiedKnowledge12.6% ±1.086 / 91
SWE-bench Verified · SWE-bench/experimentsCoding11.4%62 / 65
Chess PuzzlesReasoning4.0% ±2.087 / 90
Artificial Analysis Humanity's Last ExamKnowledge2.8%221 / 222
SWE-bench (full test split)Coding7.1%5 / 7

Sources

BenchLMCC BY-NC 4.0Epoch AICC BYSWE-bench teamNo licence stated

History

5 releases
Index23.8 → 23.3
23 Oct 202612 Nov 2026
Output / 1M$75.00 → $75.00
23 Oct 202612 Nov 2026

Questions

How much does Claude 3 Opus cost?
$15.00 per 1M input tokens, $75.00 per 1M output tokens.
How good is Claude 3 Opus?
Its index is 23.3 out of 100. That is place 274 of 283 ranked models.
What is Claude 3 Opus best at?
Knowledge: 27.6 out of 100, place 247 of 268.
What is the context window of Claude 3 Opus?
200K tokens.
Is Claude 3 Opus open source?
No. Its weights are not published.
When was Claude 3 Opus released?
March 2024.

Badge

Claude 3 Opus on Inference360: place 274 of 283, index 23.3
HTML
<a href="https://inference360.ai/models/claude-3-opus"><img src="https://inference360.ai/badges/claude-3-opus.svg" alt="Claude 3 Opus on Inference360: place 274 of 283, index 23.3" height="20"></a>
Markdown
[![Claude 3 Opus on Inference360: place 274 of 283, index 23.3](https://inference360.ai/badges/claude-3-opus.svg)](https://inference360.ai/models/claude-3-opus)

Compare with

Claude 3.5 Haiku24.1Phi-422.4Claude 3 Haiku22.0

Lower price

dots3-note Preview63.4 · FreeApodex 1.1 Mini55.7 · FreeNemotron 3 Ultra50.8 · Free

More from Anthropic

All models
Claude 3 Haiku22.0Claude 3.5 Sonnet32.2Claude 3.5 Haiku24.1Claude 3.7 Sonnet Thinking41.4Claude 3.7 Sonnet37.7