DeepSeek: DeepSeek V4 Pro 0813
deepseek/deepseek-v4-pro-0813DeepSeek V4 Pro 0813 is a large-scale mixture-of-experts model from DeepSeek. This is the GA release of DeepSeek V4 Pro.
Modalities
In / out per 1M$0.66 / $1.98
Context1M
ReleasedAug 12, 2026
Providers
23 providers serve this model upstream.
| Provider | Input /M | Output /M | Cache read /M | Latency P50 | Throughput P50 | Uptime |
|---|---|---|---|---|---|---|
| $0.90 | $0.58 | $0.19 | 1.47s | 73 tps | 100% | |
| $0.90 | $0.58 | $0.19 | 1.54s | 75 tps | 100% | |
| $0.66 | $1.98 | $0.02 | 1.97s | 45 tps | 100% | |
 SiliconFlowfp8 | $1.32 | $3.96 | $0.04 | 2.19s | 54 tps | 100% |
| $0.96 | $2.88 | $0.10 | 3.73s | 47 tps | 100% | |
| $1.06 | $3.17 | $0.04 | 2.89s | 38 tps | 100% | |
 DigitalOcean | $1.32 | $3.96 | $0.04 | 0.81s | 43 tps | 100% |
| $1.32 | $3.96 | $0.13 | 1.48s | 61 tps | 100% | |
| $1.32 | $3.96 | $0.04 | 1.12s | 45 tps | 100% | |
 DeepSeek | $0.66 | $1.98 | $0.02 | 1.24s | 28 tps | 100% |
| $0.37 | $2.93 | $0.09 | 0.99s | 59 tps | 99.97% | |
| $0.99 | $2.97 | $0.03 | 1.72s | 41 tps | 99.97% | |
| $0.25 | $5.00 | $0.25 | 0.66s | 115 tps | 99.92% | |
 DeepInfraus · fp8 | $1.30 | $2.60 | $0.10 | 1.50s | 50 tps | 99.89% |
| $1.32 | $3.96 | $0.04 | 1.95s | 35 tps | 99.67% | |
| $1.32 | $3.96 | $0.13 | 0.77s | 108 tps | 99.66% | |
| $1.32 | $3.96 | $0.13 | 1.35s | 44 tps | 99.33% | |
| $1.32 | $3.96 | $0.04 | 1.12s | 40 tps | 99.25% | |
| $1.12 | $3.37 | $0.11 | 1.41s | 61 tps | 99.1% | |
| $0.88 | $3.80 | $0.15 | 1.57s | 68 tps | 98.62% | |
| $1.65 | $4.95 | $0.17 | 1.28s | 47 tps | 98.48% | |
| $0.23 | $8.00 | $0.23 | 1.33s | 43 tps | 97.69% | |
 CoreWeavefp8 | $1.31 | $3.96 | $0.04 | 0.56s | 115 tps | 96.45% |
Market-wide measurements, refreshed every 15 minutes. They describe who serves DeepSeek V4 Pro 0813 in general, not how ProjectCOZY routes your requests.
Pricing
List rates per 1M tokens — what DeepSeek V4 Pro 0813 costs pay-as-you-go elsewhere, not what you pay here.
Input
$0.66
per 1M tokens
Output
$1.98
per 1M tokens
Cache read
$0.02
per 1M cached tokens
Included in every pass
DeepSeek V4 Pro 0813 is included in every ProjectCOZY pass at no extra cost — one flat rate, no per-token metering. The rates above are there for comparison.
Performance
Time to first token and output speed across every provider, over the last 3 days.
0.0s20.0s40.0s60.0s80.0s
Oct 16, 12 AMOct 16, 8 AMOct 16, 4 PMOct 17, 12 AMOct 17, 8 AMOct 17, 4 PMOct 18, 12 AMOct 18, 8 AMOct 18, 4 PMOct 19, 12 AMOct 19, 8 AM
Sail Research (us) Ionstream Together Relace DeepSeek SiliconFlow GMICloud Cloudflare Venice Phala Wafer Sail Research DigitalOcean Alibaba Cloud Int. AtlasCloud NovitaAI DeepInfra (us) CoreWeave Parasail StreamLake Baidu Qianfan Alibaba Cloud Int. (us-east-1) DekaLLM
0 tok/s500 tok/s1000 tok/s1500 tok/s2000 tok/s
Oct 16, 12 AMOct 16, 8 AMOct 16, 4 PMOct 17, 12 AMOct 17, 8 AMOct 17, 4 PMOct 18, 12 AMOct 18, 8 AMOct 18, 4 PMOct 19, 12 AMOct 19, 8 AM
Sail Research (us) Ionstream Together Relace DeepSeek SiliconFlow GMICloud Cloudflare Venice Phala Wafer Sail Research DigitalOcean Alibaba Cloud Int. AtlasCloud NovitaAI DeepInfra (us) CoreWeave Parasail StreamLake Baidu Qianfan Alibaba Cloud Int. (us-east-1) DekaLLM
Uptime
100% of requests were served successfully over the last 3 days, across all providers.
Last 3 daysAvailability 99.66%
Oct 16, 11:00 AMNow
Availability over the last 24 hoursAvailability 99.65%
100% 98% 97% 95%
Oct 18, 11:40 AMNow
Apps
Tools that most often run DeepSeek V4 Pro 0813, by tokens this week.
Activity
4359.3B tokens processed market-wide in the last month.
0100.0B200.0B300.0B400.0B
Sep 19Sep 24Sep 29Oct 4Oct 9Oct 14Oct 19
Tokens
FAQ
Which model id do I use for DeepSeek V4 Pro 0813?
deepseek/deepseek-v4-pro-0813. Send it as the model field from any OpenAI-compatible client, against http://192.168.1.55/v1.
How much context does DeepSeek V4 Pro 0813 hold?
1,048,576 tokens per request, of which up to 384,000 can be the response.
What can it accept and return?
Input: text. Output: text.
What does DeepSeek V4 Pro 0813 cost on a pass?
Nothing beyond the pass itself. Pay-as-you-go elsewhere it lists at $0.66 per 1M input tokens and $1.98 per 1M output tokens.
Who built DeepSeek V4 Pro 0813?
DeepSeek, in the DeepSeek family.
Explore
Other models your pass can call.
