DeepSeek V4 Pro vs Kimi K3

Q: DeepSeek V4 Pro 和 Kimi K3 哪个更便宜？

在 输入 / 1m tokens 方面，DeepSeek V4 Pro 更便宜（$1.608 对比 $3，相差 1.9×）。其他行可能得出相反的结论——上方表格提供了完整信息，实际成本取决于你的组合使用情况。

更换任一侧 vs

何时使用哪一个 — 经过整理的结论，而非基准测试表格

两者都是接受文本输入的推理模型，带对话、代码和工具，但 deepseek-v4-pro 是更便宜的纯文本路径：每百万输入 $1.608、输出 $3.216，对比 kimi-k3 的 $3 和 $15——输入约 1.9 倍、输出约 4.7 倍——而且它 $0.0134 的缓存读取比 kimi-k3 的 $0.3 便宜约 22 倍。需要图像或视频输入、1048576 token 上下文，或单次回复最多 1048576 输出 token 时选 kimi-k3；大批量文本工作、1000000 token 上下文与最多 393216 输出，以及关闭思考的选项（kimi-k3 不允许）则选 deepseek-v4-pro。

定价

	DeepSeek V4 Pro	Kimi K3	Δ
输入 / 1M tokens	$1.608	$3	0.54×
输出 / 1M tokens	$3.216	$15	0.21×
缓存读取 / 1M tokens	$0.0134	$0.3	0.045×
缓存写入	不单独收费	—	—

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 63 个聊天模型的每 1M token 的输入价格（对数刻度）

DeepSeek V4 Pro · $1.608 Kimi K3 · $3

$0.05 · Qwen3 VL Flash $30 · GPT-5.4 Pro

能力

	DeepSeek V4 Pro	Kimi K3
工具使用	是	是
思考控制	可配置	始终开启
结构化输出	是	是
提示词缓存	隐式（自动）	隐式（自动）
缓存生存时间	no fixed TTL (evicted when unused)	未公开
最小缓存前缀	未公开	未公开

规格

	DeepSeek V4 Pro	Kimi K3
输入模态	文本	文本图像视频
输出模态	文本	文本
发布日期	2026-04-24	—
上下文窗口	1M	1M
最大输出	393K	1M
思考参数	thinking.type reasoning_effort	reasoning_effort (top-level; the thinking object is not accepted)
允许的值	thinking.type enabled disabled reasoning_effort high max (low and medium map to high, xhigh maps to max)	reasoning_effort low high max
默认值	enabled, with reasoning_effort high some complex agent requests are automatically set to max	max

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： DeepSeek V4 Pro · Kimi K3

单个 Prompt，两个模型 —— 通过网关实测

面向 CFO 的推理修复含 bug 的函数提取至 schema 按约束条件写作

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

DeepSeek V4 Pro 通过 · 3 sentences

Enabling reasoning causes the model to generate a lengthy internal chain‑of‑thought before it delivers a visible answer, and we are billed for every one of those hidden reasoning tokens just as we are for the final output. In most interactions, the volume of these unseen tokens at least equals the visible response, effectively doubling the total tokens consumed per query. That doubled token consumption translates directly into a doubled per‑query cost, which is why our overall LLM bill doubled.

输出 641 tok (+545 思考) 延迟 11.5 s

Kimi K3 通过 · 3 sentences

Reasoning models don't just answer questions—they "think" first, generating long internal chains of step-by-step logic before producing a response. Those hidden thinking steps are billed as output tokens (the most expensive kind, typically 3–5x the price of input tokens), and a single query can generate thousands of them even when the visible answer is only a paragraph long. So you're paying for dramatically more compute per request: the bill doubled because the model does far more work behind the scenes, not because usage increased.

输出 755 tok (+637 思考) 延迟 20.8 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

DeepSeek V4 Pro 通过 · 8/8 cases

```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 best = max(best, cur) else: cur = 1 return best ``` **Explanation:** The bug stems from initializing `cur = 0` and only incrementing it when two adjacent elements match, which counts the number of *additional* equal elements rather than the actual run length (which is `cur + 1`). As a result, the returned `best` is always one less than the true longest run—most obviously returning 0 for a single-element list instead of 1. […]

输出 2418 tok (+2131 思考) 延迟 35.6 s

Kimi K3 通过 · 8/8 cases

```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 best = max(best, cur) else: cur = 1 return best ``` The bug is a fencepost error: `cur` counts *matching adjacent pairs* rather than *items in the run*, because it starts at 0 and only increments when `xs[i] == xs[i - 1]`. A run of length L contains L−1 equal pairs, so every run was undercounted by exactly one. For a single-element list (a run of length 1) the loop never runs and the function returns 0 instead of 1; likewise any run ending at the last index — e.g. the two `2`s in `[1, 2, 2]` — reported 1 instead of 2, making it look like no run existed at all. […]

输出 1837 tok (+1547 思考) 延迟 47.2 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

DeepSeek V4 Pro 通过 · 5/5 fields, guidance null

{ "company": "Meridian Robotics", "quarter": "September quarter", "revenue_usd_m": 412, "growth_pct": "eleven and change", "guidance": null }

输出 1204 tok (+1153 思考) 延迟 19.9 s

Kimi K3 通过 · 5/5 fields, guidance null

{ "company": "Meridian Robotics", "quarter": "September quarter", "revenue_usd_m": 412, "growth_pct": 11, "guidance": null }

输出 924 tok (+863 思考) 延迟 25.9 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

DeepSeek V4 Pro 没有可供评分的回答 · no answer text within 16,384 tokens (all of it went to thinking)

模型未返回回答文本 —— 整个 token 预算都用在了隐藏思考上。

输出 8193 tok (+8192 思考) 延迟 106.2 s

Kimi K3 通过 · 120 words, 0 banned, 1 question

Meet Prompt Cache, a new API gateway capability that stores prompt responses and serves them across OpenAI, Anthropic, Google, and Azure endpoints. It matches requests by model, prompt hash, tools, temperature, and tenant policy, so repeated work returns fast while sensitive contexts stay isolated. Teams set TTLs, stale rules, encryption scopes, and bypass flags per route. Analytics show hit rate, latency saved, token spend avoided, and drift risk by provider. What changes for developers? Keep one integration, add cache headers, and watch fallback logic respect consent, residency, and audit needs. During rollout, canary keys compare fresh answers with cached copies before promotion. Prompt Cache cuts vendor calls, steadies p95 latency, and gives platform owners controls for cost, quality, and compliance.

输出 1527 tok (+1354 思考) 延迟 37.9 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="deepseek-v4-pro",
    # model="kimi-k3",  # 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)

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://synthorai.io/v1",
  apiKey: "sk-syn-...",
});

const resp = await client.chat.completions.create({
  model: "deepseek-v4-pro",
  // model: "kimi-k3",  // 取消注释此行，注释上一行
  messages: [{ role: "user", content: "Summarize this diff" }],
  reasoning_effort: "medium",
});
console.log(resp.choices[0].message.content);

curl https://synthorai.io/v1/chat/completions \
  -H "Authorization: Bearer sk-syn-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-pro",
    # "model": "kimi-k3",  # 取消注释此行，注释上一行
    "messages": [{"role": "user", "content": "Hello"}],
    "reasoning_effort": "medium"
  }'

package main

import (
	"context"
	"fmt"

	"github.com/openai/openai-go/v3"
	"github.com/openai/openai-go/v3/option"
)

func main() {
	client := openai.NewClient(
		option.WithBaseURL("https://synthorai.io/v1"),
		option.WithAPIKey("sk-syn-..."),
	)
	resp, _ := client.Chat.Completions.New(context.TODO(), openai.ChatCompletionNewParams{
		Model: "deepseek-v4-pro",
		// Model: "kimi-k3",  // 取消注释此行，注释上一行
		Messages: []openai.ChatCompletionMessageParamUnion{
			openai.UserMessage("Summarize this diff"),
		},
		ReasoningEffort: openai.ReasoningEffortMedium,
	})
	fmt.Println(resp.Choices[0].Message.Content)
}

import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.models.chat.completions.*;
import com.openai.models.ReasoningEffort;

OpenAIClient client = OpenAIOkHttpClient.builder()
    .baseUrl("https://synthorai.io/v1")
    .apiKey("sk-syn-...")
    .build();

ChatCompletion resp = client.chat().completions().create(
    ChatCompletionCreateParams.builder()
        .model("deepseek-v4-pro")
        // .model("kimi-k3")  // 取消注释此行，注释上一行
        .addUserMessage("Summarize this diff")
        .reasoningEffort(ReasoningEffort.MEDIUM)
        .build());
System.out.println(resp.choices().get(0).message().content().orElse(""));

获取 API 密钥 →

常见问题

DeepSeek V4 Pro 和 Kimi K3 哪个更便宜？

在输入 / 1m tokens 方面，DeepSeek V4 Pro 更便宜（$1.608 对比 $3，相差 1.9×）。其他行可能得出相反的结论——上方表格提供了完整信息，实际成本取决于你的组合使用情况。

我可以在不进行两次集成的情况下，对 DeepSeek V4 Pro 和 Kimi K3 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。

DeepSeek V4 Pro 和 Kimi K3 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。

来自我们的实测研究

DeepSeek V4 Pro 正式版对比预览版：思考量减少 18-62%一张图算多少 token？同一图标实测 15 个 API：6 到 1,298 LLM 思考控制实测：13 个模型如何接受、忽略或执行参数

定价

能力

规格

单个 Prompt，两个模型 —— 通过网关实测

只需一行代码即可在它们之间切换

常见问题

相关对比

来自我们的实测研究