DeepSeek V4 Pro (Reasoning, High Effort)
Intelligence
47.3
#27/523
Coding
33.1
#40/429
Math
—
Speed
29 tok/s
TTFT: 1.42s
Pricing
$1.74 / $3.48
per 1M tokens (in/out)
DeepSeek V4 Pro (Reasoning, High Effort) is DeepSeek’s advanced model designed for complex reasoning tasks. It processes at 29.286 tokens per second and is priced at $1.74 per million input tokens, targeting professional users in data-intensive fields.
When to Use DeepSeek V4 Pro (Reasoning, High Effort)
✓ Best For
- Complex reasoning tasks in AI applications.
- Data analysis requiring high effort and accuracy.
- Professional environments needing efficient token processing.
✗ Not Ideal For
- Users seeking high-speed performance for simple tasks.
- Applications with low budget constraints.
How DeepSeek V4 Pro (Reasoning, High Effort) Compares
Intelligence Index · Higher is better
OpenAIAlibabaDeepSeekZ AIAnthropic
Benchmark Profile
Coding Index
MiniMaxZ AIDeepSeekOpenAIAnthropic
Output Speed · tok/s
DeepSeekXiaomiMetaAnthropic
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (9)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 47.3 | |
| Coding Index | 33.1 | |
| GPQA | 905% | |
| HLE | 335% | |
| SciCode | 46.4% | |
| IFBench | 71.3% | |
| LCR | 65% | |
| TerminalBench Hard | 41.7% | |
| Tau2 | 94.2% |
Data: Artificial Analysis · Updated: April 24, 2026
Frequently Asked Questions (15)
When was DeepSeek V4 Pro (Reasoning, High Effort) released?
DeepSeek V4 Pro (Reasoning, High Effort) was released on April 24, 2026.
Who created DeepSeek V4 Pro (Reasoning, High Effort)?
DeepSeek V4 Pro (Reasoning, High Effort) was created by DeepSeek.
How intelligent is DeepSeek V4 Pro (Reasoning, High Effort)?
DeepSeek V4 Pro (Reasoning, High Effort) scores 47 on the Artificial Analysis Intelligence Index, placing it well above average among other open weight models of similar size (median: 28).
How verbose is DeepSeek V4 Pro (Reasoning, High Effort)?
When evaluated on the Intelligence Index, DeepSeek V4 Pro (Reasoning, High Effort) generated 100M output tokens, which is at the higher end compared to other open weight models of similar size (median: 43M).
Is DeepSeek V4 Pro (Reasoning, High Effort) a reasoning model?
Yes, DeepSeek V4 Pro (Reasoning, High Effort) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
What input modalities does DeepSeek V4 Pro (Reasoning, High Effort) support?
DeepSeek V4 Pro (Reasoning, High Effort) supports text input.
What output modalities does DeepSeek V4 Pro (Reasoning, High Effort) support?
DeepSeek V4 Pro (Reasoning, High Effort) supports text output.
Can DeepSeek V4 Pro (Reasoning, High Effort) process images?
No, DeepSeek V4 Pro (Reasoning, High Effort) does not support image input. It can only process text.
Is DeepSeek V4 Pro (Reasoning, High Effort) multimodal?
No, DeepSeek V4 Pro (Reasoning, High Effort) is not multimodal. It only supports text input.
What is the context window of DeepSeek V4 Pro (Reasoning, High Effort)?
DeepSeek V4 Pro (Reasoning, High Effort) has a context window of 1.0M tokens. This determines how much text and conversation history the model can process in a single request.
Is DeepSeek V4 Pro (Reasoning, High Effort) open source?
Yes, DeepSeek V4 Pro (Reasoning, High Effort) is open weights. The model weights are publicly available and can be downloaded for self-hosting.
How many parameters does DeepSeek V4 Pro (Reasoning, High Effort) have?
DeepSeek V4 Pro (Reasoning, High Effort) has 1.6 trillion parameters (49 billion active).
What are the active parameters of DeepSeek V4 Pro (Reasoning, High Effort)?
DeepSeek V4 Pro (Reasoning, High Effort) is a Mixture of Experts (MoE) model with 1.6 trillion total parameters, but only 49 billion active parameters are used during inference.
What is the license for DeepSeek V4 Pro (Reasoning, High Effort)?
DeepSeek V4 Pro (Reasoning, High Effort) is released under the MIT license. This license allows commercial use.
How does DeepSeek V4 Pro (Reasoning, High Effort) perform on benchmarks?
DeepSeek V4 Pro (Reasoning, High Effort) achieves a score of 47 on the Artificial Analysis Intelligence Index. This composite benchmark evaluates models across reasoning, knowledge, mathematics, and coding.
