Claude 4 Opus (Reasoning)
Intelligence
39
#94/523
Coding
34
#102/429
Math
73.3
#86/265
Speed
37 tok/s
TTFT: 6.56s
Pricing
$15.00 / $75.00
per 1M tokens (in/out)
Claude 4 Opus (Reasoning) is Anthropic’s AI model designed for advanced reasoning tasks. It processes at 36.755 tokens per second and is priced at $15 per million input tokens and $75 per million output tokens, targeting professional users.
When to Use Claude 4 Opus (Reasoning)
✓ Best For
- Complex problem-solving in professional environments.
- Data analysis and interpretation.
- Software development and coding assistance.
✗ Not Ideal For
- Basic conversational tasks.
- High-speed real-time applications.
How Claude 4 Opus (Reasoning) Compares
Intelligence Index · Higher is better
OpenAIXiaomiAnthropic
Benchmark Profile
Coding Index
TencentAnthropicOpenAIGoogle
Output Speed · tok/s
OpenAIMicrosoftAnthropicMistralKimi
Math Index
Z AIAllen Institute for AIAnthropicGoogleAlibaba
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (15)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 39 | |
| Coding Index | 34 | |
| Math Index | 73.3 | |
| MMLU-Pro | 873% | |
| GPQA | 796% | |
| LiveCodeBench | 636% | |
| HLE | 117% | |
| SciCode | 39.8% | |
| IFBench | 53.7% | |
| LCR | 33.7% | |
| TerminalBench Hard | 31.1% | |
| Tau2 | 73.4% | |
| AIME | 75.7% | |
| AIME 2025 | 73.3% | |
| MATH 500 | 98.2% |
Data: Artificial Analysis · Updated: April 10, 2026
Frequently Asked Questions (15)
When was Claude 4 Opus (Reasoning) released?
Claude 4 Opus (Reasoning) was released on May 22, 2025.
Who created Claude 4 Opus (Reasoning)?
Claude 4 Opus (Reasoning) was created by Anthropic.
How intelligent is Claude 4 Opus (Reasoning)?
Claude 4 Opus (Reasoning) scores 39 (estimated) on the Artificial Analysis Intelligence Index, placing it above average among other reasoning models in a similar price tier (median: 31).
How fast is Claude 4 Opus (Reasoning)?
Claude 4 Opus (Reasoning) generates output at 31.8 tokens per second (based on Anthropic's API), which is at the lower end compared to other reasoning models in a similar price tier (median: 67.8 t/s).
What is the latency of Claude 4 Opus (Reasoning)?
Claude 4 Opus (Reasoning) has a time to first token (TTFT) of 11.03s (based on Anthropic's API), which is at the higher end compared to other reasoning models in a similar price tier (median: 2.57s).
How much does Claude 4 Opus (Reasoning) cost?
Claude 4 Opus (Reasoning) costs $15.00 per 1M input tokens (at the higher end, median: $1.35) and $75.00 per 1M output tokens (at the higher end, median: $8.40), based on Anthropic's API.
What is Claude 4 Opus (Reasoning) API pricing?
Claude 4 Opus (Reasoning) costs $15.00 per 1M input tokens and $75.00 per 1M output tokens (based on Anthropic's API). For a blended rate (3:1 input to output ratio), this is $30.00 per 1M tokens. Pricing may vary by provider.
Is Claude 4 Opus (Reasoning) a reasoning model?
Yes, Claude 4 Opus (Reasoning) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
What input modalities does Claude 4 Opus (Reasoning) support?
Claude 4 Opus (Reasoning) supports text and image input.
What output modalities does Claude 4 Opus (Reasoning) support?
Claude 4 Opus (Reasoning) supports text output.
Can Claude 4 Opus (Reasoning) process images?
Yes, Claude 4 Opus (Reasoning) supports image input and can analyze, describe, and answer questions about images.
Is Claude 4 Opus (Reasoning) multimodal?
Yes, Claude 4 Opus (Reasoning) is multimodal. It can process text and image input and generate text output.
What is the context window of Claude 4 Opus (Reasoning)?
Claude 4 Opus (Reasoning) has a context window of 200k tokens. This determines how much text and conversation history the model can process in a single request.
Is Claude 4 Opus (Reasoning) open source?
No, Claude 4 Opus (Reasoning) is proprietary. The model weights are not publicly available.
How many parameters does Claude 4 Opus (Reasoning) have?
Claude 4 Opus (Reasoning) is a proprietary model and Anthropic has not disclosed the model size or parameter count.
