Deepseek deepseek-v4-flash

DeepSeek-V4-Flash is DeepSeek's cost-efficient open-source model, released on April 24, 2026. It is a 284B parameter Mixture-of-Experts (MoE) language model with only 13B active parameters, pre-trained on 32T tokens, supporting a context length of one million tokens. V4-Flash delivers reasoning performance approaching V4-Pro while being significantly faster and cheaper — making it ideal for high-volume, latency-sensitive workloads.

---

Why It Looks Great

• Mixture-of-Experts architecture with 284B total parameters and only 13B active — the smallest activation among Tier-1 models
• 1000000 context window powered by Compressed Sparse Attention (CSA) and DeepSeek Sparse Attention (DSA)
• Near V4-Pro reasoning performance at a fraction of the cost

---

Key Features

• Context Window: 1000000 tokens
• Max Output: 384000 tokens
• Vision: Not Supported
• Function Calling: Supported
• Thinking Mode: Supported (non-thinking / high / max)
• JSON Output: Supported
• FIM Completion: Supported (non-thinking mode only)

Benchmarks

Specifications

---

How to Use

1. Write your prompt — describe the task, provide context, and specify desired output format.
2. Submit — the model processes your request and returns the response.

API Integration

Base URL: https://llm.wavespeed.ai/v1 API Endpoint: chat/completions Model ID: deepseek/deepseek-v4-flash

---

API Usage

Python SDK

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://llm.wavespeed.ai/v1"
)

response = client.chat.completions.create(
    model="deepseek/deepseek-v4-flash",
    messages=[{"role": "user", "content": "Hello!"}]
)

print(response.choices[0].message.content)

cURL

curl https://llm.wavespeed.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "deepseek/deepseek-v4-flash",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

---

Notes

• Model: deepseek/deepseek-v4-flash
• Provider: Deepseek
• Open-source weights available on HuggingFace and ModelScope
• Supports both OpenAI and Anthropic API formats
• For simple Agent tasks, V4-Flash performs on par with V4-Pro; for complex agentic workflows, consider V4-Pro