01Overview02Models03Rankings04Compare05Quiz06Guide

Model

Claude Opus 4.7 NO.03

Previous-generation frontier Opus, still highly reliable for agentic coding.

CompareBack to the index

Anthropic · Claude · Frontier · Closed weights

Specification

Fig.01 — Claude — Anthropic

USA

Context window
1Mtokens
Max output
128Ktokens
Input price
$5/ 1M
Output price
$25/ 1M
Cached input
$0.50/ 1M
Throughput
—
Class
Frontier
Modalities
Text, Image, Pdf
Weights
Closed
Knowledge cutoff
Nov 2025
Released
Mar 2026

Overview

Claude Opus 4.7 is the previous-generation frontier Opus, superseded by 4.8 but still a genuinely frontier-reliable model for long-horizon agentic coding. It remains a strong choice for teams standardized on it that value stability over chasing the newest release. Same pricing and context as 4.8 with slightly lower peak capability.

Strengths

  • Frontier-grade agentic reliability
  • Strong real-world coding
  • Large 1M-token context
  • Well-proven, mature behavior in production

Trade-offs

  • Superseded by Opus 4.8 at identical pricing
  • Slightly lower peak capability than the current frontier
  • No audio/video input

Fit

Best for

  • Existing 4.7-standardized pipelines
  • Long-horizon coding agents
  • Complex reasoning tasks
  • Baseline comparisons against 4.8

Not ideal for

  • New deployments (prefer Opus 4.8)
  • Cost-sensitive high-volume use

Capabilities

Ten axes, normalized 0–100 and scored the same way across the whole catalog — so a 78 here means what a 78 means anywhere else on the bench.

Fig.02 — Capability radar

0 — 100

Reasoning
94
Coding
94
Math
90
Writing
92
Knowledge
91
Speed
55
Agentic
94
Vision
86
Multilingual
89
Long Context
93

Benchmarks

Public results, with independent third-party runs marked. Bars normalize percentages against 100 and Elo ratings against a 1500 ceiling.

Independent index

Artificial Analysis Intelligence Index

Composite of ~9–10 independent evals · Artificial Analysis

54/ 100

BenchmarkResult
Artificial Analysis Intelligence IndexIndependentGeneral · Artificial Analysis
54%
SWE-bench VerifiedIndependentCoding · SWE-bench
83.5%
SWE-bench ProIndependentCoding · SWE-bench Pro
64.3%
DeepSWEIndependentCoding · Datacurve
54%
GPQA DiamondIndependentReasoning · Independent aggregators
94.2%
LiveBenchIndependentGeneral · LiveBench
76.9%
LMArena EloIndependentGeneral · LMArena
1505Elo

Alternatives

Models in roughly the same class — the ones worth weighing against this record.

Compare side by side
01Claude Opus 4.8AnthropicState-of-the-art autonomous coding reliability at half the price of Fable 5.02GPT-5.6 SolOpenAIOpenAI's GPT-5.6 flagship — its best coding-and-agentic model yet, for the hardest work.03Claude Fable 5AnthropicAnthropic's most capable model — the reliability frontier for the hardest agentic work.

Index

PreviousClaude Opus 4.8NextClaude Sonnet 4.6
All models/EDUCATION