DeepSeek V4 Pro
Chatdeepseek/deepseek-v4-proDeepSeek V4 Pro is DeepSeek's large-scale Mixture-of-Experts flagship, released on 2026-04-24, with 1.6T total parameters and 49B activated parameters. It is built for advanced reasoning, coding, and long-horizon agent workflows, and delivers strong performance across knowledge, math, and software engineering tasks. The model supports tool use (function calling) and prompt caching, so long system prompts and shared context stay inexpensive across turns. Context window: 1M tokens, output: 384K. Pricing is $0.45/M input tokens and $0.88/M output tokens, with cached input reads at $0.0037/M. Available via OpenAI and Anthropic protocols through Ofox.
Providers
/v1/chat/completions/v1/responses/v1/chat/completionsCode Examples
from openai import OpenAIclient = OpenAI(base_url="https://api.ofox.run/v1",api_key="YOUR_OFOX_API_KEY",)response = client.chat.completions.create(model="deepseek/deepseek-v4-pro",messages=[{"role": "user", "content": "Hello!"}],)print(response.choices[0].message.content)
Uptime & Status
Benchmarks
DeepSeek V4 Pro scores 1457 in the Overall category of the LMArena text leaderboard (style control), ranking #42 of 374 models based on 41,800 human preference votes (updated 2026-07-12).
| Category | Arena Score | 95% CI | Rank | Votes |
|---|---|---|---|---|
| Overall | 1453–1462 | #42 of 374 | 41,800 | |
| Hard Prompts | 1474–1485 | #42 of 374 | 27,599 | |
| Coding | 1496–1509 | #46 of 369 | 12,300 | |
| Math | 1431–1457 | #53 of 362 | 2,278 | |
| Creative Writing | 1435–1452 | #36 of 372 | 6,799 | |
| Instruction Following | 1446–1459 | #38 of 374 | 14,048 | |
| Chinese | 1492–1521 | #26 of 344 | 2,049 |
Source: LMArena · CC BY 4.0 · Updated 2026-07-12 · Methodology ↗ · Ranks compare models within each category of the LMArena text leaderboard (style control). Scores come from third-party human preference evaluations, not from OFOX.