Opus 4.7vsGPT-5.4
在 18 个共同 benchmark 中，Opus 4.7 整体领先：Opus 4.7 领先 10 项，GPT-5.4 领先 8 项，持平 0 项，平均分差 +7.49。
Opus 4.710 项(56%)(44%)8 项GPT-5.4
评测分数
按能力类目分组，每组内按分差大小排列；共 18 项。
Math and Reasoning
GPT-5.4 领先 5/5| 评测项 | Opus 4.7 | GPT-5.4 | 分差 |
|---|---|---|---|
| MathArena Apex | 40.629 / 17Extra-High (With Tools) | 54.178 / 17Extra-High (With Tools) | -13.55 |
| FrontierMath - Tier 4 | 22.9012 / 80Extra-High (No Tools) | 27.1011 / 80Extra-High (No Tools) | -4.20 |
| FrontierMath | 43.806 / 60Extra-High (No Tools) | 47.605 / 60Extra-High (No Tools) | -3.80 |
| HMMT Feb 2026 | 93.946 / 10Extra-High (With Tools) | 97.732 / 10Extra-High (With Tools) | -3.79 |
| AIME 2026 | 95.8310 / 29Extra-High (With Tools) | 99.174 / 29Extra-High (With Tools) | -3.34 |
General Knowledge
Opus 4.7 领先 2/3| 评测项 | Opus 4.7 | GPT-5.4 | 分差 |
|---|---|---|---|
| HLE | 33.30184 / 563Normal (No Tools) · Text only | 11.30374 / 563Normal (No Tools) · Text only | +22 |
| CritPt | 5.1094 / 200Normal (No Tools) | 0.60168 / 200Normal (No Tools) | +4.50 |
| LiveBench | 76.538 / 117Deep Thinking (No Tools) | 77.975 / 117Deep Thinking (No Tools) | -1.44 |
Agent Level Benchmark
Opus 4.7 领先 2/2| 评测项 | Opus 4.7 | GPT-5.4 | 分差 |
|---|---|---|---|
| Terminal Bench Hard | 54.5015 / 244Normal (With Tools) | 37.9059 / 244Normal (With Tools) | +16.60 |
| τ²-Bench - Telecom | 74128 / 264Normal (With Tools) | 64.30151 / 264Normal (With Tools) | +9.70 |
Claw-style Agent Evaluation
Opus 4.7 领先 1/1| 评测项 | Opus 4.7 | GPT-5.4 | 分差 |
|---|---|---|---|
| PinchBench v2 | 76.0115 / 45Reported best (effort unspecified) | 75.7017 / 45Reported best (effort unspecified) | +0.31 |
Coding and Software Engineer
Opus 4.7 领先 1/1| 评测项 | Opus 4.7 | GPT-5.4 | 分差 |
|---|---|---|---|
| WeirdML v2 | 76.4013 / 52Normal (With Tools) | 57.4431 / 52Normal (With Tools) | +18.96 |
Instruction Following
GPT-5.4 领先 1/1| 评测项 | Opus 4.7 | GPT-5.4 | 分差 |
|---|---|---|---|
| IF Bench | 43.60189 / 282Normal (No Tools) | 48.40164 / 282Normal (No Tools) | -4.80 |
Long Context
Opus 4.7 领先 1/1| 评测项 | Opus 4.7 | GPT-5.4 | 分差 |
|---|---|---|---|
| AA-LCR | 75.7082 / 170Normal (No Tools) | 58.30130 / 170Normal (No Tools) | +17.40 |
Multimodal Understanding
Opus 4.7 领先 1/1| 评测项 | Opus 4.7 | GPT-5.4 | 分差 |
|---|---|---|---|
| MMMU-Pro | 76.4073 / 227Normal (No Tools) | 70.60123 / 227Normal (No Tools) | +5.80 |
Text Embedding
GPT-5.4 领先 1/1| 评测项 | Opus 4.7 | GPT-5.4 | 分差 |
|---|---|---|---|
| Context Arena | 22.12122 / 126Normal (No Tools) | 32.79112 / 126Normal (No Tools) | -10.67 |
Writing and Creative Capabilities
Opus 4.7 领先 1/1| 评测项 | Opus 4.7 | GPT-5.4 | 分差 |
|---|---|---|---|
| Creative Writing | 1,9079 / 106Normal (No Tools) | 1,83614 / 106Normal (No Tools) | +71.50 |
科学与综合推理
Opus 4.7 领先 1/1| 评测项 | Opus 4.7 | GPT-5.4 | 分差 |
|---|---|---|---|
| GPQA Diamond | 88.50110 / 462Normal (No Tools) | 74.80284 / 462Normal (No Tools) | +13.70 |
规格对比
| 字段 | Opus 4.7 | GPT-5.4 |
|---|---|---|
| 发布机构 | Anthropic | OpenAI |
| 发布时间 | 2026-04-16 | 2026-03-05 |
| 模型类型 | 推理大模型 | 多模态大模型 |
| 架构 | 稠密模型 | 稠密模型 |
| 参数规模 | 暂无数据 | 暂无数据 |
| 上下文长度 | 1000K | 1M |
| 最大输出 | 128K | 125K |
API 调用价格
价格优先使用 DataLearner 配置的 API 记录；缺失项不做推测。
| 价格项 | Opus 4.7 | GPT-5.4 |
|---|---|---|
| 文本输入 | $5 / 1M tokens | $2.5 / 1M tokens |
| 文本输出 | $25 / 1M tokens | $15 / 1M tokens |
| 缓存读取 | $0.5 / 1M tokens | $0.25 / 1M tokens |
| 缓存写入 | $6.25 / 1M tokens | 暂无公开价格 |
小结
- Opus 4.7在以下类目领先:General Knowledge (2/3)、Agent Level Benchmark (2/2)、Claw-style Agent Evaluation (1/1)、Coding and Software Engineer (1/1)、Long Context (1/1)、Multimodal Understanding (1/1)、Writing and Creative Capabilities (1/1)、科学与综合推理 (1/1)
- GPT-5.4在以下类目领先:Math and Reasoning (5/5)、Instruction Following (1/1)、Text Embedding (1/1)
18 个共同 benchmark 上，Opus 4.7 平均高出 7.49 分。
单项差距最大的 benchmark：Creative Writing — Opus 4.7 1,907，GPT-5.4 1,836（分差 +71.50）。
本页正文由结构化模型、价格与 benchmark 数据生成，不使用实时 LLM 撰写。