Evalry
All models

Anthropic: Claude Opus 4.5

by Anthropic

Claude Opus 4.5 is Anthropic’s frontier reasoning model optimized for complex software engineering, agentic workflows, and long-horizon computer use. It offers strong multimodal capabilities, competitive performance across real-world coding and reasoning benchmarks, and improved robustness to prompt injection. The model is designed to operate efficiently across varied effort levels, enabling developers to trade off speed, depth, and token usage depending on task requirements. It comes with a new parameter to control token efficiency, which can be accessed using the OpenRouter Verbosity parameter with low, medium, or high. Opus 4.5 supports advanced tool use, extended context management, and coordinated multi-agent setups, making it well-suited for autonomous research, debugging, multi-step planning, and spreadsheet/browser manipulation. It delivers substantial gains in structured reasoning, execution reliability, and alignment compared to prior Opus generations, while reducing token overhead and improving performance on long-running tasks.

Avg Score

85.2%

235 answers

Avg Latency

4.0s

15 runs

Pricing

$5.00

input

/

$25.00

output

per 1M tokens

Context

200K

tokens

Alternatives

Models with similar or better quality but different tradeoffs

Same Quality, Cheaper

Models with similar or better performance at a lower cost per token.

ModelCost
Xiaomi: MiMo-V2-Flash (free)-100%
Google: Gemma 3 12B-100%
Mistral: Pixtral 12B-98%
Z.AI: GLM 4 32B -98%
Google: Gemma 3 12B-98%

Same Quality, Faster

Models with similar or better performance but lower latency.

ModelLatency
Google: Gemini 3 Flash Preview-54%
OpenAI: GPT-4.1 Nano-54%
Nous: Hermes 4 70B-26%
OpenAI: GPT-5.1 Chat-20%
Anthropic: Claude Sonnet 4.5-15%

Same Cost, Better

Models at a similar price point with higher benchmark scores.

ModelScore
Mistral: Pixtral 12B+5%
Baidu: ERNIE 4.5 VL 424B A47B +5%
AllenAI: Olmo 3 32B Think+5%
Nous: Hermes 4 70B+5%
Google: Gemma 3 12B+5%

Other Models from Anthropic

Compare performance with other models from the same creator

ModelScoreLatencyCost/1M
Anthropic: Claude Sonnet 4.582.6%4.0s$9.00
Anthropic: Claude 3.7 Sonnet80.0%8.9s$9.00
Anthropic: Claude Haiku 4.577.8%2.8s$3.00
Anthropic: Claude 3.5 Haiku76.6%3.4s$2.40
Anthropic: Claude 3 Haiku74.3%1.8s$0.75
Anthropic: Claude 3.7 Sonnet50.0%5.5s$9.00
Anthropic: Claude 3.5 Sonnet50.0%3.0s$18.00
Anthropic: Claude Opus 4.10.0%13.1s$45.00
Anthropic: Claude Opus 40.0%12.4s$45.00
Anthropic: Claude Sonnet 40.0%6.5s$9.00
Anthropic: Claude 3.5 Haiku (2024-10-22)——$2.40

Benchmark Performance

How this model performs across different benchmarks

BenchmarkScoreRank
Math Bench: Addition (1-10)
100.0%
6 / 35
Math Bench: Subtraction (1-10)
100.0%
6 / 35
Math Bench: Multiplication (1-10)
100.0%
6 / 35
Math Bench: Division (1-10)
100.0%
6 / 35
Math Bench: Arithmetic (1-100)
100.0%
13 / 35
Math Bench: Arithmetic (1-1000)
100.0%
6 / 35
Character Frequency Bench
100.0%
1 / 35
Motor production test
100.0%
6 / 36
Spatial Reasoning: Germany
95.5%
3 / 35
Product Recommendations
95.0%
1 / 36
Karlsruhe Local Knowledge Benchmark
87.2%
7 / 35
Westat Test
50.0%
223 / 291
Money Boy Cultural Literacy Test
45.7%
11 / 35
German Memelord Bench
42.1%
6 / 35
Karl Lorey knowlege
2.7%
19 / 36

Price vs Performance

Compare cost efficiency across all models

Current model
Other models
X-axis uses log scale for better visualization of price range

Score Over Time

Performance trends across all benchmark runs

Benchmark Activity

Number of benchmark runs over time

Quickstart

Get started with this model using OpenRouter

View on OpenRouter
import { OpenRouter } from "@openrouter/sdk";

const openrouter = new OpenRouter({
  apiKey: "<OPENROUTER_API_KEY>"
});

const completion = await openrouter.chat.completions.create({
  model: "anthropic/claude-opus-4.5",
  messages: [
    {
      role: "user",
      content: "Hello!"
    }
  ]
});

console.log(completion.choices[0].message.content);

Get your API key at openrouter.ai/keys