DeepSeek-V4.1-FlashvsClaude Opus 5
在 13 个共同 benchmark 中，Claude Opus 5 整体领先：DeepSeek-V4.1-Flash 领先 5 项，Claude Opus 5 领先 8 项，持平 0 项，平均分差 -5.07。
DeepSeek-V4.1-Flash
DeepSeek-AI · 2026-09-10 · 多模态大模型
Claude Opus 5
Anthropic · 2026-07-24 · 推理大模型
DeepSeek-V4.1-Flash5 项(38%)(62%)8 项Claude Opus 5
评测分数
按能力类目分组，每组内按分差大小排列；共 13 项。
AI Agent - Tool Usage
Claude Opus 5 领先 2/3| 评测项 | DeepSeek-V4.1-Flash | Claude Opus 5 | 分差 |
|---|---|---|---|
| Terminal-Bench 4.0 | 31.207 / 16Max (With Tools) | 51.803 / 16Max (With Tools) | -20.60 |
| Terminal-Bench 3.0 | 304 / 11Max (With Tools) | 43.301 / 11Max (With Tools) | -13.30 |
| Terminal-Bench 2.1 | 90.601 / 53Max (With Tools) | 89.103 / 53Max (With Tools) | +1.50 |
Coding and Software Engineer
Claude Opus 5 领先 2/3| 评测项 | DeepSeek-V4.1-Flash | Claude Opus 5 | 分差 |
|---|---|---|---|
| Program Bench | 20.308 / 11Max (With Tools) | 375 / 11Max (With Tools) | -16.70 |
| NL2Repo-Bench | 65.402 / 16Max (With Tools) | 75.301 / 16Max (With Tools) | -9.90 |
| DeepSWE | 74.202 / 38Max (With Tools) | 744 / 38Max (With Tools) | +0.20 |
Multimodal Understanding
Claude Opus 5 领先 3/3| 评测项 | DeepSeek-V4.1-Flash | Claude Opus 5 | 分差 |
|---|---|---|---|
| Chartography | 78.904 / 8Max (With Tools) | 842 / 8Max (With Tools) | -5.10 |
| BabyVision | 89.603 / 9Max (With Tools) | 94.101 / 9Max (With Tools) | -4.50 |
| ZeroBench Main | 493 / 6Max (With Tools) | 522 / 6Max (With Tools) | -3 |
Agent Level Benchmark
DeepSeek-V4.1-Flash 领先 1/1| 评测项 | DeepSeek-V4.1-Flash | Claude Opus 5 | 分差 |
|---|---|---|---|
| Agents' Last Exam | 31.806 / 19Max (With Tools) | 28.607 / 19Max (With Tools) | +3.20 |
General Knowledge
DeepSeek-V4.1-Flash 领先 1/1| 评测项 | DeepSeek-V4.1-Flash | Claude Opus 5 | 分差 |
|---|---|---|---|
| HLE | 63.903 / 197Max (With Tools) | 63.604 / 197Max (With Tools) | +0.30 |
Productivity Knowledge
DeepSeek-V4.1-Flash 领先 1/1| 评测项 | DeepSeek-V4.1-Flash | Claude Opus 5 | 分差 |
|---|---|---|---|
| AutomationBench | 54.801 / 17Max (With Tools) | 50.303 / 17Max (With Tools) | +4.50 |
科学与综合推理
Claude Opus 5 领先 1/1| 评测项 | DeepSeek-V4.1-Flash | Claude Opus 5 | 分差 |
|---|---|---|---|
| GPQA Diamond | 90.9039 / 274Max (No Tools) | 93.4018 / 274Max (No Tools) | -2.50 |
规格对比
| 字段 | DeepSeek-V4.1-Flash | Claude Opus 5 |
|---|---|---|
| 发布机构 | DeepSeek-AI | Anthropic |
| 发布时间 | 2026-09-10 | 2026-07-24 |
| 模型类型 | 多模态大模型 | 推理大模型 |
| 架构 | MoE 架构 | 稠密模型 |
| 参数规模 | 5520亿 | 暂无数据 |
| 上下文长度 | 1M | 1M |
| 最大输出 | 384K | 128K |
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 | DeepSeek-V4.1-Flash | Claude Opus 5 |
|---|---|---|
| 文本输入 | ¥1 / 1M tokens | $5 / 1M tokens |
| 文本输出 | ¥4 / 1M tokens | $25 / 1M tokens |
| 缓存读取 | ¥0.02 / 1M tokens | $0.5 / 1M tokens |
| 缓存写入 | 暂无公开价格 | $6.25 / 1M tokens |
小结
- DeepSeek-V4.1-Flash在以下类目领先:Agent Level Benchmark (1/1)、General Knowledge (1/1)、Productivity Knowledge (1/1)
- Claude Opus 5在以下类目领先:AI Agent - Tool Usage (2/3)、Coding and Software Engineer (2/3)、Multimodal Understanding (3/3)、科学与综合推理 (1/1)
13 个共同 benchmark 上，Claude Opus 5 平均高出 5.07 分。
单项差距最大的 benchmark：Terminal-Bench 4.0 — DeepSeek-V4.1-Flash 31.20，Claude Opus 5 51.80（分差 -20.60）。
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。