Start building

DeepSeek V4 Flash

deepseek/deepseek-v4-flash

DeepSeek V4 Flash is an efficiency-optimized Mixture-of-Experts model from DeepSeek with 284B total parameters and 13B activated parameters, supporting a 1M-token context window. It is designed for fast inference and high-throughput workloads, while maintaining strong reasoning and coding performance. The model includes hybrid attention for efficient long-context processing. Reasoning efforts `high` and `xhigh` are supported; `xhigh` maps to max reasoning. It is well suited for applications such as coding assistants, chat systems, and agent workflows where responsiveness and cost efficiency are important.

Use in agentGet API key
Input price
$0.44USD / 1M tokens
Output price
$1.32USD / 1M tokens
Context window
N/A
Max output
384K

Try the model

Playground

Open playground
Input⌘ / Ctrl + Enter

Try a prompt

1M context384K max output
Input $0.44/M · Output $1.32/M · Cache read $0.01/M
OutputReady

Your response will appear here

Choose an example or write a prompt, then click Run.

Specifications

Pricing

Input
$0.44 / 1M tokens
Output
$1.32 / 1M tokens
Billing formula
usage.prompt_tokens * 0.44 / 1000000 + usage.completion_tokens * 1.32 / 1000000 + usage.cached_tokens * 0.000008 / 1000000

Context & modalities

Max output
384,000 tokens
Input
Text
Output
Text

Capabilities

Chat
Supported
Vision
Not supported
Reasoning
Supported
Structured output
Supported
Function calling
Supported
Audio input
Not supported

Access

Provider
DeepSeek
Model ID
deepseek/deepseek-v4-flash
Execution
sync
API
Chat Completions API
Endpoint
/v1/chat/completions

Start building

Send your first request

OpenAI-compatible endpoint with unified authentication and usage tracking.

Production API
Chat Completions API endpoint
https://api.sandbase.ai/v1/chat/completions
Model ID
deepseek/deepseek-v4-flash
# The quoted heredoc keeps Unicode and shell metacharacters unchanged.
result=$(curl --fail-with-body --silent \
  -X POST "https://api.sandbase.ai/v1/chat/completions" \
  -H "Authorization: Bearer $SANDBASE_API_KEY" \
  -H "Content-Type: application/json" \
  --data-binary @- <<'SANDBASE_JSON'
{
  "model": "deepseek/deepseek-v4-flash",
  "messages": [
    {
      "role": "user",
      "content": "Hello"
    }
  ],
  "max_tokens": 2048
}
SANDBASE_JSON
)
printf '%s\n' "$result"

Choose your model

Compare models

All language models
ModelContextInput / 1MOutput / 1MReleased
DeepSeek: DeepSeek V4 FlashThis model
DeepSeek		—$0.44$1.32Apr 24, 2026
DeepSeek: DeepSeek V4.1 Flash
DeepSeek		1M——Sep 10, 2026
DeepSeek: DeepSeek V4 Flash Vision Exp (Experimental)
DeepSeek		1M——Aug 21, 2026
DeepSeek: DeepSeek V4 Flash 0731
DeepSeek		1M——Jul 31, 2026
DeepSeek: DeepSeek V4 Pro
DeepSeek		1M——Apr 24, 2026
DeepSeek: DeepSeek V3.2 Speciale
DeepSeek		160K——Dec 1, 2025