DataLearner logo

Opus 4.7vsOpus 4.5

Across 12 shared benchmarks, Opus 4.7 leads overall: Opus 4.7 wins 11, Opus 4.5 wins 1, with 0 ties and an average score difference of +30.51.

Anthropic
Opus 4.7

Anthropic · 2026-04-16 · Reasoning model

Anthropic
Opus 4.5

Anthropic · 2025-11-25 · Reasoning model

Opus 4.711 wins(92%)(8%)1 winOpus 4.5

Benchmark scores

Grouped by capability, sorted by largest gap within each. 12 shared benchmarks.

Coding and Software Engineer

Opus 4.7 3/3
BenchmarkOpus 4.7Opus 4.5Diff
Text Arena (Coding)1,5627 / 35Normal (No Tools)1,47920 / 35Normal (No Tools)+83.39
GSO44.121 / 21Normal (With Tools)26.508 / 21Normal (With Tools)+17.62
SWE-bench Verified87.606 / 116Extended (with tools)80.909 / 116Extended (with tools)+6.70

Agent Level Benchmark

Even 2/2
BenchmarkOpus 4.7Opus 4.5Diff
Terminal Bench Hard54.5015 / 244Normal (With Tools)40.9050 / 244Normal (With Tools)+13.60
τ²-Bench - Telecom74128 / 264Normal (With Tools)86.3082 / 264Normal (With Tools)-12.30

AI Agent - Tool Usage

Opus 4.7 1/1
BenchmarkOpus 4.7Opus 4.5Diff
Terminal Bench 2.069.406 / 48Extended (with tools)59.3020 / 48Extended (with tools)+10.10

General Evaluation

Opus 4.7 1/1
BenchmarkOpus 4.7Opus 4.5Diff
GPQA Diamond88.50110 / 462Normal (No Tools)81220 / 462Normal (No Tools)+7.50

General Knowledge

Opus 4.7 1/1
BenchmarkOpus 4.7Opus 4.5Diff
CritPt5.1094 / 200Normal (No Tools)0.30182 / 200Normal (No Tools)+4.80

Instruction Following

Opus 4.7 1/1
BenchmarkOpus 4.7Opus 4.5Diff
IF Bench43.60189 / 282Normal (No Tools)43194 / 282Normal (No Tools)+0.60

Long Context

Opus 4.7 1/1
BenchmarkOpus 4.7Opus 4.5Diff
AA-LCR75.7082 / 170Normal (No Tools)70.7099 / 170Normal (No Tools)+5

Multimodal Understanding

Opus 4.7 1/1
BenchmarkOpus 4.7Opus 4.5Diff
MMMU-Pro76.4073 / 228Normal (No Tools)71.20121 / 228Normal (No Tools)+5.20

Writing and Creative Capabilities

Opus 4.7 1/1
BenchmarkOpus 4.7Opus 4.5Diff
Creative Writing1,9079 / 106Normal (No Tools)1,68329 / 106Normal (No Tools)+223.90

Specs

FieldOpus 4.7Opus 4.5
PublisherAnthropicAnthropic
Release date2026-04-162025-11-25
Model typeReasoning modelReasoning model
ArchitectureDenseDense
ParametersNot availableNot available
Context length1000K200K
Max output128K64K

API pricing

Prices use DataLearner records when available; missing fields are not inferred.

ItemOpus 4.7Opus 4.5
Text input$5 / 1M tokens$5 / 1M tokens
Text output$25 / 1M tokens$25 / 1M tokens
Cache read$0.5 / 1M tokens$0.5 / 1M tokens
Cache write$6.25 / 1M tokens$6.25 / 1M tokens

Summary

  • Opus 4.7leads in:Coding and Software Engineer (3/3), AI Agent - Tool Usage (1/1), General Evaluation (1/1), General Knowledge (1/1), Instruction Following (1/1), Long Context (1/1), Multimodal Understanding (1/1), Writing and Creative Capabilities (1/1)
  • Tied in:Agent Level Benchmark

On average across the 12 shared benchmarks, Opus 4.7 scores 30.51 higher.

Largest single-benchmark gap: Creative Writing — Opus 4.7 1,907 vs Opus 4.5 1,683 (+223.90).

Page generated from structured model, pricing and benchmark records. No real-time LLM is used to write the prose.