---
title: "DeepSeek V4 Pro API: facts and benchmarks | Kendr"
canonical: "https://kendr.org/models/kc-deepseek-v4-pro"
date_modified: "2026-08-12"
profile_kind: "hosted"
---

# DeepSeek V4 Pro API model profile

DeepSeek V4 Pro is a DeepSeek model profile with a 1,048,576-token context window. The dated catalog snapshot records text modalities and interfaces for agentic, coding, prompt caching, reasoning, reasoning controls.

DeepSeek V4 Pro is available under the Kendr alias kc-deepseek-v4-pro. Current account availability and customer credit quotes come from Kendr's live public model API and applicable account policy.

DeepSeek V4 Pro accepts text and returns text. The snapshot records agentic, coding, prompt caching, reasoning, reasoning controls, structured output, text generation, tool calling, and tools as capabilities or interfaces.

The dated reference snapshot lists $1.168 input and $2.336 output per million tokens. Cached input is $0.09855 per million tokens. The separately published provider-route reference is Provider-specific (DeepSeek $0.435 / $0.87; QwenCloud $2.40 / $4.80 ($0.20/M cached)).

The 2026-08-12 snapshot includes 3 Artificial Analysis indexes and 11 Design Arena categories. Publisher-reported launch results are labeled and are not presented as Kendr measurements; scores from different suites are not treated as interchangeable.

DeepSeek V4 Pro ranked #7 in the cited trailing-7-day third-party catalog usage dataset as of 2026-08-12. Observed within the cited trailing-7-day third-party catalog usage dataset; this is not global AI market share.

Profile data was reviewed for the 2026-08-12 snapshot. Claims retain their source dates, and unavailable fields are shown as unavailable instead of estimated.

## Profile facts

- **Profile type:** Kendr API model
- **Kendr API alias:** kc-deepseek-v4-pro
- **Model developer:** DeepSeek
- **Kendr route provider:** Provider-routed DeepSeek
- **Context window:** 1M
- **Snapshot date:** 2026-08-12
- **Reference catalog ID:** deepseek/deepseek-v4-pro

## Context, modalities, and identity

DeepSeek V4 Pro accepts text and returns text. The snapshot records agentic, coding, prompt caching, reasoning, reasoning controls, structured output, text generation, tool calling, and tools as capabilities or interfaces.

- **Model developer:** DeepSeek
- **Kendr route provider:** Provider-routed DeepSeek
- **Context window:** 1M
- **Maximum output:** 393,216 tokens
- **Input modalities:** text
- **Output modalities:** text
- **Knowledge cutoff:** Not disclosed

## Capabilities and supported controls

Capabilities and parameters are reported from the dated catalog and source records; their presence does not guarantee identical behavior across every provider route.

- **Supported parameters:** frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p

- agentic
- coding
- prompt caching
- reasoning
- reasoning controls
- structured output
- text generation
- tool calling
- tools

## Dated pricing snapshot

The dated reference snapshot lists $1.168 input and $2.336 output per million tokens. Cached input is $0.09855 per million tokens. The separately published provider-route reference is Provider-specific (DeepSeek $0.435 / $0.87; QwenCloud $2.40 / $4.80 ($0.20/M cached)).

Reference figures are dated 2026-08-12; the live Kendr quote can differ by selected provider route, context tier, caching, tools, region, and current rate card.

- **Price date:** 2026-08-12
- **Input:** $1.168 per 1M tokens
- **Cached input:** $0.09855 per 1M tokens
- **Output:** $2.336 per 1M tokens

## Benchmark evidence and limitations

The 2026-08-12 snapshot includes 3 Artificial Analysis indexes and 11 Design Arena categories. Publisher-reported launch results are labeled and are not presented as Kendr measurements; scores from different suites are not treated as interchangeable.

Third-party catalog fields normalized from the cited model dataset; benchmark methodology and coverage differ by source.

- **Intelligence Index:** 45.3
- **Coding Index:** 59.4
- **Agentic Index:** 37.8

| Design Arena category | Elo | Win rate | Rank |
| --- | --- | --- | --- |
| fullstack | 948 | 22.1% | #35 |
| godotgamedev | 1059 | 34% | #27 |
| webapps | 1000 | 26.4% | #34 |
| 3d | 1309 | 57.6% | #12 |
| asciiart | 1189 | 46.8% | #27 |
| codecategories | 1268 | 53.2% | #27 |
| dataviz | 1225 | 49.4% | #40 |
| gamedev | 1271 | 54.4% | #24 |
| svg | 1181 | 46.1% | #38 |
| uicomponent | 1256 | 51.5% | #32 |
| website | 1256 | 51.8% | #32 |

## Popularity and market context

DeepSeek V4 Pro ranked #7 in the cited trailing-7-day third-party catalog usage dataset as of 2026-08-12. Observed within the cited trailing-7-day third-party catalog usage dataset; this is not global AI market share.

- **Third-party catalog rank:** #7
- **Global market share:** Not inferred
- **Observation date:** 2026-08-12

## OpenAI-compatible API example

```bash
curl https://api.kendr.org/v1/chat/completions \
  -H "Authorization: Bearer $KENDR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"kc-deepseek-v4-pro","messages":[{"role":"user","content":"Hello"}]}'
```

## Frequently asked questions

### What is DeepSeek V4 Pro?

DeepSeek V4 Pro is a DeepSeek model profile with a 1,048,576-token context window. The dated catalog snapshot records text modalities and interfaces for agentic, coding, prompt caching, reasoning, reasoning controls.

### What context window does DeepSeek V4 Pro have?

The dated profile lists 1M of context and up to 393,216 output tokens. Provider routes, variants, and runtime configuration can impose lower effective limits.

### What benchmark evidence is available for DeepSeek V4 Pro?

The 2026-08-12 snapshot includes 3 Artificial Analysis indexes and 11 Design Arena categories. Publisher-reported launch results are labeled and are not presented as Kendr measurements; scores from different suites are not treated as interchangeable. Third-party catalog fields normalized from the cited model dataset; benchmark methodology and coverage differ by source.

### What is the Kendr API alias for DeepSeek V4 Pro?

Use kc-deepseek-v4-pro as the model value. The live public model API and signed-in catalog remain authoritative for route availability and customer credit quotes.

### How is DeepSeek V4 Pro priced on Kendr?

The dated reference snapshot lists $1.168 input and $2.336 output per million tokens. Cached input is $0.09855 per million tokens. The separately published provider-route reference is Provider-specific (DeepSeek $0.435 / $0.87; QwenCloud $2.40 / $4.80 ($0.20/M cached)). Kendr applies one 5% markup to configured provider model cost; the live quote and settled routing receipt are authoritative for a request.

## Sources and evidence dates

- [DeepSeek V4 Pro third-party catalog record](https://openrouter.ai/deepseek/deepseek-v4-pro) — catalog, checked 2026-08-12
- [Third-party model catalog methodology](https://openrouter.ai/docs/guides/overview/models) — methodology, checked 2026-08-12
- [DeepSeek official model documentation](https://api-docs.deepseek.com/quick_start/pricing/) — primary, checked 2026-08-12
- [Artificial Analysis capability indices methodology](https://artificialanalysis.ai/methodology/capability-indices) — benchmark, checked 2026-08-12
- [Design Arena leaderboard and methodology](https://www.designarena.ai/leaderboard) — benchmark, checked 2026-08-12
- [DeepSeek V4 Pro provider or route reference](https://www.qwencloud.com/models/deepseek-v4-pro) — primary, checked 2026-08-12

Live Kendr operational availability is published separately at https://api.kendr.org/api/public/models.