DeepSeek V4 Pro 0813
deepseek/deepseek-v4-pro-0813- DeepSeek
- Family · DeepSeek V4 Pro
- Open weights
- Text → Text
- Reasoning · adjustable
- Tools · JSON
Not on OSMEXO yet
Spec & list price
Context1M
Max output944K
Input / 1M$0.12
Output / 1M$4.0034.8× input
Usage
Tokens / day80.7B▼ 18% 7d
Usage rank#36
Requests / day3.9M
Tool calls / req0.17
Per request · 30 days
Avg input20.0K1.3× median
Avg output5130.7× median
Cache hit84%1.2× median
Reasoning share61%about median
Tool-call failures1.1%0.6× median
Consumption
Peak 254.3B 8 OctLow 59.2B 1 NovRank today #36
News & trends
11 mentions this week ▲ 22%DeepSeek V4 Pro family · videos, posts, articles · 30 days
Trending · 14 days · DeepSeek V4 Pro family
- 我用最新的 256GB Mac 跑 1.6 兆參數 DeepSeek V4 PRO825 views
- NEW DeepSeek V4 Pro Jailbreak That Actually Works!139 views
- Deepseek V4 pro is now the most expensive model in history of company247 upvotes
News · latest from DeepSeek
- Tell HN: Mistrap Pro includes DeepSeek v4.1 Flash 1017docs.mistral.ai · 1 Nov
- Show HN: DeepSeek v4.1 Flash-1017 Safety Evaluationtwitter.com · 26 Oct
- DeepSeek V4 Pro 0813 vs Claude Fable 5 on DeepSWE: Cost, Coding, and RoutingTogether AI · 17 Aug
- DeepSeek-V4 Flash 0731 vs GPT-5.6 Luna on DeepSWE: Cost and CodingTogether AI · 6 Aug
Benchmarks
Benchmark data: Artificial AnalysisOutput speed#265/33443 tok/s1.6× slower than median
Fastest provider122 tok/sbest of 26 providers
First token#1530.96 sabout median
Family progress
Intelligence Index
- Family releases36.0
Technical details
- Reasoning
- on · high
- Tokenizer
- DeepSeek
- License
- MIT
- Parameters
- 1.7T
- Moderation
- none
- Downloads · likes
- 152.1K · 909
Reasoning effort
- max
- high · default
- low
- none
Request parameters 20
- tools
- tool_choice
- response_format
- structured_outputs
- reasoning
- reasoning_effort
- include_reasoning
- temperature
- top_p
- top_k
+10 more
- min_p
- max_tokens
- stop
- seed
- frequency_penalty
- logit_bias
- logprobs
- presence_penalty
- repetition_penalty
- top_logprobs
When to use
Best for
- Hard reasoning and coding tasks over very long inputs
- Agents that call tools and return structured output
- Large-output generation such as long reports
- Self-hosted deployments (open weights)
Think twice for
- Image input (text only)
- High-volume simple chat (a Flash tier costs less)
- Latency-sensitive interactive features
Research papers
Same family
DeepSeek V4 Pro- 36.0
- 30.4
Want DeepSeek V4 Pro 0813 on OSMEXO?
Sign in to request it. Requests are reviewed; a request does not guarantee availability.