मुख्य सामग्री पर जाएं

DeepSeek V4 Pro 0813 (Reasoning, Max Effort)

DeepSeekDeepSeekओपन वेटMIT · व्यावसायिक उपयोग

विवरण

DeepSeek-V4-Pro-0813 is the GA / production release of DeepSeek-V4-Pro, with major upgrades for agentic coding and tool-use workloads. It is served behind the official API id `deepseek-v4-pro` and is distinct from catalog `deepseek-v4-pro-max`, which is the maximum reasoning-effort mode of Pro rather than this dated GA build. Same Pro MoE scale as peers (1.6T total / 49B activated). It supports low, high, and max reasoning effort, a 1M-token context window, up to 384K output tokens, and DeepSeek's OpenAI-compatible Chat Completions and Responses APIs. GA open weights are on Hugging Face at `deepseek-ai/DeepSeek-V4-Pro-0813` (MIT), distinct from the April preview repo `deepseek-ai/DeepSeek-V4-Pro`. New time-of-day pricing takes effect August 16: peak rates are $0.044 cached input, $1.32 uncached input, and $3.96 output per million tokens; off-peak rates are half those amounts.

रिलीज़ तिथि
2026-08-13
पैरामीटर
1.6T
संदर्भ लंबाई
1.0M
मोडैलिटीज़
text

क्षमता रडार

51
general
66
coding
93
reasoning
67
science
50
agents
0
multimodal

रैंकिंग

बेंचमार्क स्कोर (LLM Stats)

(LLM Stats (zeroeval))

Agents

Terminal-Bench 2.187.9%स्वयं
BrowseCompOpenAI (2025)83.4%स्वयं
CyberGym83.3%स्वयं
MCP Atlas73.6%स्वयं
DSBench-FullStack71.1%स्वयं
Terminal-Bench 2.0Stanford × Laude Institute (2026)67.9%स्वयं
DSBench-Hard67.2%स्वयं
DeepSWE62.7%स्वयं
NL2Repo61.5%स्वयं
SWE-Bench ProPrinceton NLP (2024)55.4%स्वयं
Toolathlon51.8%स्वयं
AutomationBench31.8%स्वयं
FrontierSWE29.0%
Agents' Last Exam25.7%स्वयं
FrontierCode 1.117.6%

Biology

GPQANYU + Cohere + Anthropic (2023)90.1%स्वयं

Code

LiveCodeBench93.5%स्वयं
SWE-Bench Verified80.6%स्वयं
SWE-bench Multilingual76.2%स्वयं

Factuality

SimpleQA57.9%स्वयं

Finance

MMLU-Pro87.5%स्वयं

General

CSimpleQA84.4%स्वयं
MRCR 1M83.5%स्वयं
LiveBench73.6%
CorpusQA 1M62.0%स्वयं

Math

CodeForces1.00 / 3000स्वयं
HMMT Feb 2695.2%स्वयं
MathArena Apex90.2%स्वयं
IMO-AnswerBench89.8%स्वयं
Humanity's Last Exam (with tools, text-only)60.0%स्वयं
Humanity's Last Exam48.2%स्वयं
Humanity's Last Exam (no tools, text-only)42.7%स्वयं

AA मूल्यांकन सूचकांक

(Artificial Analysis)
Coding Index(Artificial Analysis)
68.8
Intelligence Index(Artificial Analysis)
53.2
Gpqa(NYU + Cohere + Anthropic (2023))
0.9
Terminalbench V2 1
0.8
Lcr(Artificial Analysis)
0.8
Scicode(UIUC + Argonne National Lab (2024))
0.5
Hle(Center for AI Safety + Scale AI (2025))
0.4
Tau Banking
0.4

LLM Stats श्रेणी स्कोर

(LLM Stats (zeroeval))
Safety
80
Code
70
Math
60
Reasoning
60
General
60
Agents
60
Vision
60
Tool Calling
50

मूल्य निर्धारण

इनपुट मूल्य$0.435 / 1M टोकन
आउटपुट मूल्य$0.87 / 1M टोकन
मिश्रित मूल्य (3:1)$0.544 / 1M टोकन
कैश पठन मूल्य$0.00363 / 1M टोकन

गति

टोकन/सेकंड64.9
पहले टोकन में देरी1.00s
पहले उत्तर में देरी31.84s

प्रदाता मूल्य रैंकिंग

प्रदाता मूल्य रैंकिंग

23 प्रदाता

सबसे सस्ता: DeepSeekसबसे महंगा: Charm Hyper
प्रदाताइनपुटआउटपुट
1DeepSeekसबसे सस्ता
$0
$0
2routing.run
$0.348
$0.696
3CrofAI
$0.35
$0.8
4NanoGPT
$0.435
$0.87
5OpenRouter
$0.435
$0.87
6OpenCode Go
$0.435
$0.87
7Alibaba (China)
$0.435
$0.87
8Vivgrid
$0.435
$0.87
9Kilo Gateway
$0.435
$0.87
10Vercel AI Gateway
$0.435
$0.87
11LLM Gateway
$0.435
$0.87
12Auriko
$0.435
$0.87
13Merge Gateway
$0.435
$0.87
14Modelis
$0.435
$0.87
15DigitalOcean
$0.87
$1.74
16UnoRouter
$0.8999
$1.7999
17Venice AI
$1.65
$3.301
18EmpirioLabs AI
$1.65
$3.3
19Cortecs
$1.73
$3.46
20OpenCode Zen
$1.74
$3.84
21FrogBot
$1.74
$3.48
22Azure
$1.74
$3.48
23Charm Hyper
$2.4
$4.8

इस मॉडल के लिए विभिन्न API प्रदाताओं के मूल्य निर्धारण की तुलना करें।

बाहरी लिंक