Claude 4 Opus Thinking vs Claude 4 Sonnet
Change modelsClaude 4 Opus ThinkingAnthropic
Claude 4 SonnetAnthropic
IndexAll categories, out of 100.
47.0 ±7.6 best
39.5 ±5.0
GapIndex points below the top.
Top
−7.5 · no clear gap
CoverageIndex weight that has results.
75%
95% best
SpeedOutput tokens per second.
—
40/s
EloLMArena rating and rank.
—
—
ReleasedMonth of release.
2025-05
2025-05
Input / 1MUS dollars per 1M input tokens.
—
$3.00
Output / 1MUS dollars per 1M output tokens.
—
$15.00
1K chats2K tokens in, 500 out, each.
—
$13.50
1K documents20K tokens in, 1K out, each.
—
$75.00
1K agent tasks200K tokens in, 20K out, each.
—
$900.00
ContextMaximum tokens in one request.
200K
200K
AgenticMulti-step tasks with tools.
51.5 best
43.9
CodingCode writing and repair.
—
52.1
ReasoningLogic problems and puzzles.
54.8 best
38.2
MultimodalTasks with images and text.
44.9 best
36.4
KnowledgeFacts and expert knowledge.
47.1 best
36.0
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
39.2 best
31.5
MathMath problems.
43.9 best
17.0
Benchmarks
6 sharedOnly Claude 4 Opus Thinking: 5Only Claude 4 Sonnet: 11
Claude 4 Opus Thinking
Claude 4 Sonnet