Alibaba
100.00%Context1M
Max output384K
p50 latency2,730 ms
Throughput63.0 t/s
Input price$1.65
Output price$3.3
Zero retentionNo trainingCaching
Model directory
Baseer model profileLanguage model
deepseek/deepseek-v4-pro
DeepSeek V4 Pro
A technical profile of this model's capabilities, pricing, and available inference providers.
Input / 1M tokens$0.66
Output / 1M tokens$1.98
Maximum context1M
Inference providers7
Price comparison
USD per one million tokens, using current provider data.
Provider latency
Last-hour p50 in milliseconds; lower is better.
Provider uptime
Reported availability over the last 24 hours; not a contractual SLA.
Inference providers
Compare current pricing, performance, and privacy across every available provider.
Azure
100.00%Context1M
Max output128K
p50 latency1,256 ms
Throughput94.5 t/s
Input price$1.74
Output price$3.48
Zero retentionNo trainingCaching
Baseten
99.92%Context1M
Max output1M
p50 latency876 ms
Throughput139.5 t/s
Input price$1.74
Output price$3.48
Zero retentionNo trainingCaching
Deepinfra
99.89%Context65.5K
Max output65.5K
p50 latency944 ms
Throughput57.5 t/s
Input price$1.74
Output price$3.48
Zero retentionNo trainingCaching
Deepseek
99.89%Context1M
Max output384K
p50 latency1,760 ms
Throughput75.0 t/s
Input price$0.66
Output price$1.98
Caching
Novita
99.97%Context1M
Max output393.2K
p50 latency1,352 ms
Throughput60.0 t/s
Input price$1.74
Output price$3.48
No trainingCaching
Togetherai
0.00%Context1M
Max output1M
p50 latency—
Throughput—
Input price$2.1
Output price$4.4
Zero retentionNo trainingCaching
| Provider | Maximum context | Max output | p50 latency | p50 throughput | 24h uptime | Input / 1M tokens | Output / 1M tokens | Privacy |
|---|---|---|---|---|---|---|---|---|
Alibaba | 1M | 384K | 2,730 ms | 63.0 t/s | 100.000% | $1.65 | $3.3 | |
Azure | 1M | 128K | 1,256 ms | 94.5 t/s | 100.000% | $1.74 | $3.48 | |
Baseten | 1M | 1M | 876 ms | 139.5 t/s | 99.921% | $1.74 | $3.48 | |
Deepinfra | 65.5K | 65.5K | 944 ms | 57.5 t/s | 99.891% | $1.74 | $3.48 | |
Deepseek | 1M | 384K | 1,760 ms | 75.0 t/s | 99.885% | $0.66 | $1.98 | |
Novita | 1M | 393.2K | 1,352 ms | 60.0 t/s | 99.968% | $1.74 | $3.48 | |
Togetherai | 1M | 1M | — | — | 0.000% | $2.1 | $4.4 |
What can this model do?
Available modalities and capabilities across inference providers.
TextAdvanced reasoningAutomatic cachingTool useTools
Reasoning options
HighXhigh
Technical profile
- Model type
- Language model
- Release date
- Apr 23, 2026
- Knowledge cutoff
- 2025-05
- Regions
- US
- API specifications
- V2, V3, V4
- Temperature control
- Supported
Model identifier
deepseek/deepseek-v4-proModel profile updated: Sep 12, 2026, 1:46 PM
Related models
More options from the same creator or model category.