新人 免费注册,送 10 次调用,最高 $1,免绑卡。

Claude Opus 4.8 vs Claude Opus 5

vs

何时用哪一个

两者费率完全相同:每百万输入 $5、输出 $25、缓存读取 $0.5,上下文同为 1000000 token,单次最多输出 128000 token,所以决定因素不是成本。claude-opus-5 从 512 token 起就能缓存,而 claude-opus-4-8 需要 1024 token;前者默认开启思考并提供 xhigh 与 max 两档 effort,后者默认关闭思考。升级改变的是行为和短前缀的缓存命中率,不是账单:预算照旧,但要重新检查你固定过的 thinking 与 effort 设置。

Benchmark 成绩

领先高于同侪均值无人分数更高Claude Opus 4.8082 / 12918 / 129Claude Opus 51414 / 147 / 14

双方都被测过的 14 项。

Claude Opus 4.8 Claude Opus 5 其他被测模型 同侪均值 无人分数更高
DeepSWE 1.1
59%
68.8%
BioMysteryBench hard
42.4%
无人分数更高 49.4%
OSWorld 2.0
55.7%
无人分数更高 70.6%
Cybergym
78.3%
N/A
HealthBench Professional
57.4%
59.8%
GDPval-AA v2 Elo · 642-1861
1593
无人分数更高 1861
Legal Agent Benchmark held-out
10.4%
11.7%
Humanity's Last Exam no tools
49.8%
56.3%
Blueprint-Bench 2
14.5%
N/A
BrowseComp
84.3%
90.8%
Video-MME (w. sub)
86%
N/A

厂商公布: Alibaba (Qwen) Anthropic ByteDance DeepSeek Google MiniMax Moonshot OpenAI Tencent Z.ai

定价

Claude Opus 4.8 Claude Opus 5 Δ
输入 / 1M tokens $5 $5 =
输出 / 1M tokens $25 $25 =
缓存读取 / 1M tokens $0.5 $0.5 =
缓存写入 1.25x (5m) / 2x (1h) 1.25x (5m) / 2x (1h) -

费率取自构建时的实时目录;每个模型页面均附有当前的费率卡。

它们的位置 — 以该计费单位计费的所有 64 个 聊天 模型的 每 1M token 的输入价格(对数刻度)

能力

Claude Opus 4.8 Claude Opus 5
工具使用
思考控制 可配置 可配置
结构化输出
提示词缓存 显式(由您标记前缀) 显式(由您标记前缀)
缓存生存时间 5m default, 1h option 5m default, 1h option
最小缓存前缀 1024 个 token 1024 个 token

规格

Claude Opus 4.8 Claude Opus 5
输入模态 文本 图像 文本 图像
输出模态 文本 文本
发布日期 2026-05-28 2026-07-24
知识截止日期 2026-01 2026-05
上下文窗口 1M 1M
最大输出 128K 128K
思考参数
  • thinking.type "adaptive"
  • output_config.effort
  • thinking.type
  • output_config.effort
允许的值
thinking.type
  • adaptive
  • disabled
effort
  • low
  • medium
  • high
  • xhigh
  • max

"enabled" returns 400

thinking.type
  • adaptive
  • disabled
effort
  • low
  • medium
  • high
  • xhigh
  • max
默认值

thinking off

effort
  • high

thinking on

effort
  • high (Claude API and Claude Code)

规格转录自各供应商的文档;若供应商未发布某项数据,则直接省略该行,而非进行推断。 完整来源: Claude Opus 4.8 · Claude Opus 5

单个 Prompt,两个模型 —— 通过网关实测

提示词 Explain to a CFO, in exactly three sentences, why our LLM bill doubled after we enabled reasoning. 检查 恰好 3 句

Claude Opus 4.8 通过 · 3 sentences

Reasoning models generate a hidden "thinking" phase before producing their final answer, and we're billed for every one of those internal tokens even though they never appear in the user-facing output. Because this deliberation often runs several times longer than the answer itself, our token consumption per request grew substantially even though our request volume stayed flat. The upside is that this trade buys us materially higher accuracy on complex tasks, so the right question isn't just cost but whether the improved output quality justifies the per-query premium for each use case.

输出 162 tok 延迟 4.2 s

Claude Opus 5 通过 · 3 sentences

When reasoning is enabled, the model generates a hidden chain of "thinking" tokens before it writes the visible answer, and those internal tokens are billed at the output rate even though no one ever sees them — commonly 3–10x more tokens than the response itself. Because output tokens are priced several times higher than input tokens, that added volume hits us at our most expensive rate, which is why the bill roughly doubled while our request count stayed flat. The spend is controllable: we can cap the reasoning budget per call, route only genuinely hard requests to reasoning mode and cheaper models for the rest, and track cost per successfully completed task rather than cost per token, since a single correct answer usually beats three cheap retries plus human cleanup.

输出 529 tok 延迟 9.0 s

指令遵循(恰好三句,可数)、受众适配(面向 CFO 的语气),以及下方 token 计量所暴露的隐藏思考计费缺口。

提示词 This function is supposed to return the longest run of consecutive equal items, but callers report it is off by one on single-element inputs and misses runs that end at the last index. Fix it and explain the bug in one paragraph. 检查 修复通过测试

Claude Opus 4.8 通过 · 8/8 cases

## Fixed Function ```python def longest_run(xs): if not xs: return 0 best = 1 cur = 1 for i in range(1, len(xs)): if xs[i] == xs[i - 1]: cur += 1 else: cur = 1 best = max(best, cur) return best ``` ## Explanation of the Bug The original code counted **transitions between equal adjacent elements** rather than the **length of the run** itself. It initialized `best` and `cur` to `0` and incremented `cur` only when a match was found, so a run of length *N* recorded a count of *N-1* (e.g., a single element `[5]` never enters the loop and returns `0` instead of `1`, and `[5, 5]` returns `1` instead of `2`)—this is the off-by-one. […]

输出 495 tok 延迟 9.2 s

Claude Opus 5 通过 · 8/8 cases

## Fixed version ```python def longest_run(xs): if not xs: return 0 best = 1 cur = 1 for i in range(1, len(xs)): if xs[i] == xs[i - 1]: cur += 1 else: cur = 1 if cur > best: best = cur return best ``` ## The bug The counters were tracking the wrong quantity: `cur` counted *matching adjacent pairs* rather than *elements in the current run*, and both counters started at `0` instead of `1`. Since a run of length *n* contains only *n − 1* adjacent equal pairs, every result came back one too small — `[1, 1]` returned 1, `[3, 3, 3]` returned 2 — and a single-element list returned 0 because the loop body never executed and `best` kept its initial value of `0`. […]

输出 1094 tok 延迟 16.1 s

修复是否真的正确(可运行)、解释的信息密度,以及在一个边界明确的任务上的 token 效率。

提示词 Extract a JSON object with fields {company, quarter, revenue_usd_m, growth_pct, guidance} from this text. Use null for anything not stated; add no other fields. 检查 合法 JSON,schema 精确

Claude Opus 4.8 通过 · 5/5 fields, guidance null

```json { "company": "Meridian Robotics", "quarter": "Q3", "revenue_usd_m": 412, "growth_pct": 11, "guidance": null } ```

输出 72 tok 延迟 1.8 s

Claude Opus 5 通过 · 5/5 fields, guidance "Withheld pending Q4 close"

```json { "company": "Meridian Robotics", "quarter": "Q3", "revenue_usd_m": 412, "growth_pct": 11, "guidance": "Withheld pending Q4 close" } ``` Notes on judgment calls: - **quarter**: "September quarter" is labeled Q3 because the text refers to an upcoming "Q4 close," implying a calendar-aligned fiscal year. No fiscal year is given. - **revenue_usd_m**: 412 is stated as constant-currency and hedged ("call it"), so it's approximate; the currency is not explicitly stated as USD. - **growth_pct**: "up eleven and change" means slightly more than 11, so 11 is a floor rather than an exact figure.

输出 539 tok 延迟 7.6 s

schema 服从度(不臆造字段)、幻觉压力(guidance 明确被暂缓给出),以及结构化输出路径的差异。

提示词 Write a 120-word product announcement for an API gateway feature that caches prompts across providers. Forbidden words: "seamless", "unlock", "game-changer", "revolutionize", "empower". Exactly one sentence must be a question. 检查 120 词,0 个禁用词

Claude Opus 4.8 通过 · 121 words, 0 banned, 1 question

**Introducing Cross-Provider Prompt Caching** Today we're launching prompt caching directly within our API gateway, designed to cut latency and reduce your inference costs. When identical or overlapping prompts hit the gateway, cached responses are served instantly instead of being routed to the underlying model provider again. What makes this different from provider-specific caching? Our system works across OpenAI, Anthropic, Google, and any custom endpoint you configure, so you maintain one consistent cache regardless of which model handles a request. You control cache duration, invalidation rules, and scope through simple configuration settings. Teams running high-volume workloads can expect meaningful savings on repeated queries and faster response times for end users. Cross-provider prompt caching is available now for all customers on paid plans.

输出 258 tok 延迟 5.3 s

Claude Opus 5 通过 · 123 words, 0 banned, 1 question

**Cross-Provider Prompt Caching** We're introducing Cross-Provider Prompt Caching in the API Gateway. Repeated system prompts, long documents, and few-shot examples are stored once at the gateway layer and reused across OpenAI, Anthropic, Google, and self-hosted models. Instead of paying full input token costs on every request, your application sends a cache reference, and the gateway rehydrates the context before forwarding it downstream. Why does that matter? Teams running high-volume agents and retrieval pipelines typically see input token spend fall 40 to 70 percent, with median latency dropping by several hundred milliseconds. Caches are scoped per project, encrypted at rest, and invalidated automatically when a prompt template changes. Enable it with a single header, and see the docs for TTL tuning and per-route controls.

输出 1593 tok 延迟 19.1 s

约束服从度(字数预算、禁用词表、唯一的那句问句)、文风指纹,以及长度控制。

只需一行代码即可在它们之间切换

两个 ID 都包含在下方的每个选项卡中 — 高亮显示的两行是唯一的修改。相同的端点,相同的密钥,相同的请求结构。

from openai import OpenAI

client = OpenAI(
    base_url="https://synthorai.io/v1",
    api_key="sk-syn-...",
)

resp = client.chat.completions.create(
    model="claude-opus-4-8",
    # model="claude-opus-5",  # 取消注释此行,注释上一行
    messages=[{"role": "user", "content": "Summarize this diff"}],
    reasoning_effort="medium",
)
print(resp.choices[0].message.content)

获取 API 密钥 →

常见问题

Claude Opus 4.8 和 Claude Opus 5 哪个更便宜?

它们列出的 输入 / 1m tokens 相同($5),因此价格不是这一项的决定因素——请参考下文的规格与能力。

我可以在不进行两次集成的情况下,对 Claude Opus 4.8 和 Claude Opus 5 进行 A/B 测试吗?

可以。两者均通过同一个兼容 OpenAI 的端点提供服务,并使用同一把 API 密钥——切换只需更改一行模型字符串,因此你可以将一部分流量路由到各个模型并直接比较账单。

Claude Opus 4.8 和 Claude Opus 5 支持提示词缓存吗?

是的 —— 两者对缓存读取的计费均低于其输入费率,因此热前缀工作负载的成本低于标价。准确的缓存读取行位于上方的定价表中。

相关对比

来自我们的实测研究