DeepSeek V4 Pro (0813) vs Kimi K3
Which one, when — curated verdict, not a benchmark table
Both are text-out reasoning models with chat, code and tool use and roughly million-token context, so the split is price versus inputs: deepseek-v4-pro-0813 runs $1.32 per million input and $3.96 output against $3 and $15 for kimi-k3, about 2.3x cheaper on input and 3.8x on output, with cache reads at $0.132 versus $0.3. Pick kimi-k3 when you need image or video input, or single responses beyond 393,216 tokens up to its 1,048,576 max output. Pick deepseek-v4-pro-0813 for text-only work at volume, and note that kimi-k3 cannot have thinking disabled.
Pricing
| DeepSeek V4 Pro (0813) | Kimi K3 | Δ | |
|---|---|---|---|
| Input / 1M tokens | $1.32 | $3 | 0.44× |
| Output / 1M tokens | $3.96 | $15 | 0.26× |
| Cache read / 1M tokens | $0.132 | $0.3 | 0.44× |
| Cache write | no separate charge | — | — |
Rates from the live catalog at build time; each model page carries the current card.
Where they sit — input price per 1M tokens across all 63 chat models on this billing unit (log scale)
Capabilities
| DeepSeek V4 Pro (0813) | Kimi K3 | |
|---|---|---|
| Tool use | yes | yes |
| Thinking control | always on | always on |
| Structured output | yes | yes |
| Prompt caching | implicit (automatic) | implicit (automatic) |
| Cache lifetime | no fixed TTL (evicted when unused) | not published |
| Minimum cached prefix | not published | not published |
Specs
| DeepSeek V4 Pro (0813) | Kimi K3 | |
|---|---|---|
| Input modalities | text | text image video |
| Output modalities | text | text |
| Released | 2026-08-13 | — |
| Context window | 1M | 1M |
| Max output | 393K | 1M |
| Thinking parameter | reasoning_effort | reasoning_effort (top-level; the thinking object is not accepted) |
| Accepted values | reasoning_effort
| reasoning_effort
|
| Default | — | max |
Specs are transcribed from each vendor’s documentation; a row a vendor does not publish is left out rather than inferred. Full sources: DeepSeek V4 Pro (0813) · Kimi K3
One prompt, both models — measured through the gateway
DeepSeek V4 Pro (0813) passed · 3 sentences
Enabling reasoning causes the model to generate a hidden chain-of-thought with many additional tokens before producing the final answer, which sharply increases compute consumption per request. Those extra reasoning tokens are billed at the same or higher rates, so total usage doubled even though the volume of final outputs stayed similar. We can control this by adjusting reasoning effort, setting token limits, or routing only complex requests through reasoning mode.
out 308 tok (+226 thinking) latency 5.1 s
Kimi K3 passed · 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.
out 755 tok (+637 thinking) latency 20.8 s
Instruction following (exactly three sentences — countable), audience shaping (CFO register), and the hidden-thinking billing gap the token meters below expose.
DeepSeek V4 Pro (0813) passed · 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 else: cur = 1 best = max(best, cur) return best ``` The bug is that `cur` was initialized to `0` and only counted adjacent matching pairs, so a run of length `n` produced `n - 1` (or `0` for a single element) instead of `n`. The code also reset to `0` on a change, when the new run should start at length `1`. Initializing `cur` and `best` to `1` fixes both the single-element case and runs ending at the last index.
out 1315 tok (+1130 thinking) latency 16.9 s
Kimi K3 passed · 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. […]
out 1837 tok (+1547 thinking) latency 47.2 s
Whether the fix is actually correct (runnable), explanation density, and token efficiency on a bounded task.
DeepSeek V4 Pro (0813) passed · 5/5 fields, guidance null
{"company":"Meridian Robotics","quarter":"September quarter","revenue_usd_m":412,"growth_pct":11.0,"guidance":null}
out 2155 tok (+2121 thinking) latency 26.0 s
Kimi K3 passed · 5/5 fields, guidance null
{ "company": "Meridian Robotics", "quarter": "September quarter", "revenue_usd_m": 412, "growth_pct": 11, "guidance": null }
out 924 tok (+863 thinking) latency 25.9 s
Schema obedience (no invented fields), hallucination pressure (guidance is explicitly withheld), and structured-output path differences.
DeepSeek V4 Pro (0813) passed · 121 words, 0 banned, 1 question
Introducing cross-provider prompt caching in our API gateway. Send a prompt once, and the gateway stores its computed prefix across Anthropic, OpenAI, and other LLM providers. Subsequent requests with the same prompt hit the cache, cutting latency and token costs while keeping outputs consistent across routing decisions and provider failovers. Teams can route identical prompts between providers without reprocessing shared context or lengthy system instructions. How much could you save on repeated prompt prefixes? The cache respects provider-specific key formats, handles TTLs automatically, and works with streaming and batch requests. Enable it with one configuration flag—no changes to your application code. Available today on all plans. Monitor cache hit rates, token savings, and provider-specific performance metrics in the live dashboard.
out 2845 tok (+2694 thinking) latency 25.5 s
Kimi K3 passed · 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.
out 1527 tok (+1354 thinking) latency 37.9 s
Constraint obedience (word budget, banned-word list, the single question), style fingerprint, and length control.
Switch between them with one line
Both ids are in every tab below — the highlighted pair of lines is the only edit. Same endpoint, same key, same request shape.
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-0813",
# model="kimi-k3", # uncomment this line, comment the one above
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-0813",
// model: "kimi-k3", // uncomment this line, comment the one above
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-0813",
# "model": "kimi-k3", # uncomment this line, comment the one above
"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-0813",
// Model: "kimi-k3", // uncomment this line, comment the one above
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-0813")
// .model("kimi-k3") // uncomment this line, comment the one above
.addUserMessage("Summarize this diff")
.reasoningEffort(ReasoningEffort.MEDIUM)
.build());
System.out.println(resp.choices().get(0).message().content().orElse(""));FAQ
Which is cheaper, DeepSeek V4 Pro (0813) or Kimi K3?
DeepSeek V4 Pro (0813) is cheaper on input / 1m tokens ($1.32 vs $3, 2.3× apart). Other rows may point the other way — the table above carries the full card, and real cost depends on your mix.
Can I A/B test DeepSeek V4 Pro (0813) against Kimi K3 without two integrations?
Yes. Both are served through the same OpenAI-compatible endpoint with one API key — switching is a one-line model-string change, so you can route a fraction of traffic to each and compare bills directly.
Do DeepSeek V4 Pro (0813) and Kimi K3 support prompt caching?
Yes — both bill cached reads below their input rate, so warm-prefix workloads cost less than the list rates suggest. The exact cache-read rows are in the pricing table above.