Opus 4.7 Benchmark Details
Opus 4.7 currently shows benchmark results led by GPQA Diamond (4 / 187, score 94.20), SWE-bench Verified (5 / 111, score 87.60), LiveBench (7 / 115, score 76.91). This page also compares it with 2 competitor models and 3 predecessor or same-series models, including performance and pricing views when available. 3 source links are attached for reference.
Benchmark Results
Benchmark Results
General Knowledge
17 evaluationsCoding and Software Engineer
2 evaluationsMath and Reasoning
2 evaluationsAI Agent - Tool Usage
4 evaluationsCompetitor Comparison
Benchmark scores for Opus 4.7 compared against top models in its class
11 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.
| Benchmark | Opus 4.7Current | GPT-5.4 | Gemini 3.1 Pro Preview |
|---|---|---|---|
ARC-AGI 综合评估 | 92.00Thinking Level · High | 93.70Standard Mode | -- |
ARC-AGI-2 综合评估 | 75.80Thinking Level · High | 77.10Standard Mode | -- |
HLE 综合评估 | 54.70Extended Thinking ｜ Tools | 52.10Thinking Level · Extra High ｜ Tools | 51.40Thinking Level · High ｜ Tools |
LiveBench 综合评估 | 76.91Deep Thinking Mode | 80.28Deep Thinking Mode | 79.93Thinking Level · High |
SWE-Bench Pro - Public 编程与软件工程 | 64.30Extended Thinking ｜ Tools | -- | 54.20Thinking Level · High ｜ Tools |
SWE-bench Verified 编程与软件工程 | 87.60Extended Thinking ｜ Tools | -- | 80.60Thinking Level · High ｜ Tools |
22.90Thinking Level · Extra High | 27.10Thinking Level · Extra High | -- | |
BrowseComp AI Agent - 信息收集 | 79.30Extended Thinking ｜ Tools | 82.70Thinking Level · Extra High ｜ Tools | 85.90Thinking Level · High ｜ Tools |
MCP-Atlas AI Agent - 工具使用 | 79.10Deep Thinking Mode ｜ Tools | 70.60Thinking Level · Extra High ｜ Tools | -- |
OSWorld-Verified AI Agent - 工具使用 | 78.00Extended Thinking ｜ Tools | 75.00Thinking Level · Extra High ｜ Tools | -- |
Terminal Bench 2.0 AI Agent - 工具使用 | 69.40Extended Thinking ｜ Tools | 75.10Thinking Level · Extra High ｜ Tools | 68.50Thinking Level · High ｜ Tools |
Standard API Pricing: Opus 4.7 vs. Peer Models
Shows standard text input and output pricing side by side for each model. If extended-context pricing exists, the chart keeps the base rate and explains the threshold below.
Source: DataLearnerAI. Standard text prices shown here use the default supplier. · USD / 1M tokens
When a context threshold exists, the charted base price only applies within these limits:
| Model | Supplier | Standard input | Standard output | Base price applies to |
|---|---|---|---|---|
Opus 4.7 | Anthropic | $5 / 1M tokens | $25 / 1M tokens | — |
GPT-5.4 | OpenAI | $2.5 / 1M tokens | $15 / 1M tokens | <= 272K |
Gemini 3.1 Pro Preview | Google Deep Mind | $2 / 1M tokens | $12 / 1M tokens | <= 200K |
Version History
How each version of the Opus 4.7 series stacks up on benchmark tests
12 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.· Click a row to view its trend chart.
| Benchmark | Opus 4.7Current | Claude Opus 4.6 | Opus 4.5 | Opus 4.1 |
|---|---|---|---|---|
ARC-AGI 综合评估 | 92.00Thinking Level · High | 92.00Extended Thinking | -- | -- |
ARC-AGI-2 综合评估 | 75.80Thinking Level · High | 66.30Extended Thinking | -- | -- |
GPQA Diamond 综合评估 | 94.20Extended Thinking | 91.31Extended Thinking | -- | 81.00Extended Thinking |
HLE 综合评估 | 54.70Extended Thinking ｜ Tools | 53.00Extended Thinking ｜ Tools | 43.20Extended Thinking ｜ Tools | -- |
LiveBench 综合评估 | 76.91Deep Thinking Mode | -- | 75.9664K | 61.8132K |
SWE-bench Verified 编程与软件工程 | 87.60Extended Thinking ｜ Tools | 80.84Extended Thinking ｜ Tools | 80.90Extended Thinking ｜ Tools | 74.50Extended Thinking ｜ Tools |
Simple Bench 常识推理 | 61.70Standard Mode | 67.60Standard Mode | 62.00Extended Thinking | -- |
FrontierMath 数学推理 | 43.80Thinking Level · Extra High | 40.70Thinking Level · High | -- | 7.20Extended Thinking |
22.90Thinking Level · Extra High | 22.90Thinking Level · High | 4.2032K | 4.20Extended Thinking | |
BrowseComp AI Agent - 信息收集 | 79.30Extended Thinking ｜ Tools | 84.00Thinking Enabled ｜ Tools | -- | -- |
MCP-Atlas AI Agent - 工具使用 | 79.10Deep Thinking Mode ｜ Tools | 76.80Deep Thinking Mode ｜ Tools | 69.80Thinking Level · High ｜ Tools | -- |
OSWorld-Verified AI Agent - 工具使用 | 78.00Extended Thinking ｜ Tools | 72.70Extended Thinking ｜ Tools | -- | -- |
Single-Benchmark Version Trend
Viewing: ARC-AGI · 综合评估
Standard API Pricing Across the Opus 4.7 Series
Shows standard text input and output pricing side by side for each model. If extended-context pricing exists, the chart keeps the base rate and explains the threshold below.
Source: DataLearnerAI. Standard text prices shown here use the default supplier. · USD / 1M tokens
When a context threshold exists, the charted base price only applies within these limits:
| Model | Supplier | Standard input | Standard output | Base price applies to |
|---|---|---|---|---|
Opus 4.7 | Anthropic | $5 / 1M tokens | $25 / 1M tokens | — |
Claude Opus 4.6 | Anthropic | $5 / 1M tokens | $25 / 1M tokens | <= 200K |
Opus 4.5 | Facebook AI研究实验室 | $5 / 1M tokens | $25 / 1M tokens | — |
Opus 4.1 | Anthropic | $15 / 1M tokens | $75 / 1M tokens | — |