DeepSeek

DeepSeek V4 Pro

Long-context reasoning model for coding agents and high-volume workflows.

LLMreasoningcodingtoolslong context

Try in Playground Open dashboard

Model IDdeepseek/deepseek-v4-pro

Statusavailable

Context1M tokens

Max output64K tokens

Pricing

per 1M tokens

Input	$1.91
Output	$3.83
Cached input	$0.190

Benchmarks

Static frontend preview data

Cost index

0.18x

vs frontier blend

p95 latency

2.0s

streaming start

Agent tasks

91.2%

workflow solve rate

API example

Use the Tokmux base URL with your key

curl https://api.tokmux.ai/v1/chat/completions \
  -H "Authorization: Bearer $TOKMUX_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek/deepseek-v4-pro",
    "messages": [
      {"role": "user", "content": "Give me a launch checklist."}
    ]
  }'

reasoningcodingtoolslong context