DeepSeek V4 Pro / チャット
Large-scale MoE model with 1.6T total parameters and 49B activated, supporting a 1M-token context window . Designed for advanced reasoning, coding, and long-horizon agent workflows . Top performance on GPQA Diamond (88.8%) and Terminal-Bench Hard (46.2%)
Playground Chat
AIモデルとの会話を開始します。何でも質問できます。
AI生成結果の正確性は異なる場合があります。
DeepSeek V4 ProDeep Reasoning with Logprobs
DeepSeek V4 Pro offers thinking toggles, high or max reasoning effort, logprobs evaluation signals, and production sampling for demanding analytical workloads.
At a Glance
V4 Pro Reasoning Stack
Analytical depth with evaluation hooks.
High or Max Effort
Dial reasoning between high and max for multi-step proofs and hard analysis.
Thinking Toggle
Enable or disable deep thinking per request depending on task complexity.
Logprob Evaluation
logprobs and top_logprobs expose token probability signals for confidence scoring.
Production Sampling
Temperature, top-p, max tokens, and stop sequences for controlled outputs.
How It Works
Deep analysis in four steps.
Set Thinking & Effort
Enable thinking and choose high or max depth.
Stream or Wait
Stream for long analyses; single-shot for batch evals.
Collect Logprobs
Use probability signals for confidence scoring.
Sample Responsibly
Tune temperature and top-p for tone control.
V4 Pro Domains
Where analytical depth is the product.
Formal Reasoning
Max effort for proofs and multi-hop analysis.
Eval Pipelines
Logprobs for confidence and regression tests.
Research Synthesis
High effort for literature and market synthesis.
Agent Decisions
Reasoned tool choice with probability signals.
Prompt Tips
Get more from V4 Pro reasoning.
High and Max effort spend tokens more usefully when you request proof paths.
Place documents before instructions for more reliable following.
Everyday Q&A is faster at high effort or with thinking off.
DeepSeek V4 Pro Quickstart
High / max reasoning with logprobs.
curl -X POST "https://api.powertokens.ai/v1/chat/completions" \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4-pro",
"messages": [
{"role": "user", "content": "Walk through a careful proof of the AM-GM inequality for two variables."}
],
"stream": true,
"temperature": 0.7
}'Technical Specifications
Confirmed parameters and runtime execution protocols.
料金詳細
このモデルの実際の課金は、API リクエストで渡される特定のパラメータに基づいて動的に計算されます。以下は具体的な組み合わせとそれに対応する料金です：
- 平日（月〜金） 09:00 - 12:00, 14:00 - 18:00 (UTC+8) の間に送信されたリクエストには、2x の価格倍率が適用されます。
| モダリティ | 入力クレジット | 出力クレジット | 入力価格 | 出力価格 | 暗黙的キャッシュヒット | 明示的キャッシュヒット | キャッシュ作成 |
|---|---|---|---|---|---|---|---|
| Standard | 660/ 1M Tokens | 1,980/ 1M Tokens | $0.660 | $1.980 | $0.022 22/ 1M Tokens | $0.022 22/ 1M Tokens | -- |
Models from the Same Channel
Explore complementary models and alternative versions from the same provider channel.
Recommended Related Models
Explore complementary video and multimodal models with your unified API key.
GLM-5.3
glm-5.3
Zhipu AI's flagship text model optimized for complex software engineering and long-horizon Agent tasks. Features a 1M-token context window with mandatory reasoning (3 levels: low/high/max). Coding capability improved 50% over GLM-5.2 on Z.ai Code Bench; scores SOTA on Terminal Bench 3.0. Excels in cybersecurity tasks (cyber vulnerability discovery)
MiniMax-M3.1-Flash
MiniMax-M3.1-Flash
Next-gen efficiency-focused multimodal model (428B MoE) for agentic workflows, coding, and computer use . Features a 1M-token context window and tunable thinking depth (low–max) . Scores 73.7% on DeepSWE v1.1, 61.7 on OSWorld V2, and 82% on CyberGym, while delivering 200+ TPS output speed at 70% lower agent-task cost than its predecessor
GLM-5.3 Flash
glm-5.3-flash
Z.ai's native multimodal model (320B-A18B) with text, image, video, and file input. Features hybrid sparse + linear attention, reducing attention compute 3x and KV cache 4.4x vs. GLM-5.3. Scores 57 on Artificial Analysis Intelligence Index—matching Claude Opus 4.8—with 1M context and 128K max output. Excels in agentic coding, vision-based Computer Use, and CAD reproduction
Qwen3.8 Flash
qwen3.8-flash
Alibaba's cost-efficient multimodal reasoning model. Supports text, image, and video inputs with text output. Features a native 1M-token context window for long documents, codebases, and agentic workflows. Excels in coding assistance, desktop interaction, chart analysis, and long-video understanding. Compatible with OpenAI/Anthropic protocols for seamless integration
Frequently Asked Questions
Everything you need to know before integrating this model.
Start Building with DeepSeek V4 Pro Today
Create an account in seconds to receive free test credits. No upfront payment or long-term contract required.