DeepSeek V4 Pro
DeepSeek V4 Pro is DeepSeek's language model. A high-capacity variant of DeepSeek's V4 large language model tuned for extended reasoning with greater compute effort.
| Benchmarks | |
|---|---|
| Intelligence Index | #72 |
| Coding Index | #44 |
| GPQA | #61 |
| HLE | #63 |
| IFBench | #20 |
| Time to First Token | #469 |
| SciCode | #63 |
| LCR | #101 |
| TerminalBench Hard | #22 |
| TAU2 | #16 |
| Output TPS | #109 |
Capabilities
Input1/5
Text✓
Image·
Audio·
Video·
PDF·
Output1/5
Text✓
Image·
Audio·
Video·
Embedding·
Capabilities0/13
Reasoning·
Adaptive Reasoning·
Function Calling·
Parallel Function Calling·
Structured Outputs·
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill·
Versions
| Version | Released | Context | Input / 1M | Output / 1M | Status |
|---|---|---|---|---|---|
| DeepSeek V4 Pro | 1.0M | $0.115 | $1.98 | Deprecated | |
| DeepSeek V4 Pro 0813 P | — | — | — | — | Available |
| DeepSeek V4 Pro | — | — | — | — | Current |
| DeepSeek V4 Pro | 1.0M | $0.103 | $1.20 | Deprecated | |
| DeepSeek V4 Pro 813 | — | 1.0M | $1.32 | $3.96 | Deprecated |
| DeepSeek V4 Pro US | — | — | — | — | Available |
Other Models
| Model | Tier | Released | Context | Input / 1M | Output / 1M |
|---|---|---|---|---|---|
| DeepSeek V4 Flash 0731 P | Flash | — | — | — | — |
| DeepSeek V4.731 Flash US | Flash | — | — | — | — |
| DeepSeek V4 420 Flash | Flash | — | — | — | — |
| DeepSeek V4.4.1 Flash | Flash | — | — | — | — |
| DeepSeek V4.1 Flash Expires Exp | Flash | 1.0M | $0.010 | $0.300 | |
| DeepSeek V4.1 Flash Fast | Flash | 1.0M | $0.010 | $0.150 | |
| DeepSeek V4.1 Flash Beta | Flash | 1.0M | — | — | |
| DeepSeek V4.1 Flash Abliterated | Flash | — | 1.0M | $0.140 | $0.500 |
| DeepSeek V4.1 Flash DST | Flash | — | — | — | — |
| DeepSeek V4.1 Flash MaaS | Flash | — | 1.0M | $0.300 | $1.20 |