GPT-5.6 is here 🎉 20% off all GPT 🎉 All July 🔥Learn more
DeepSeek

DeepSeek V4 Flash

Chat
deepseek/deepseek-v4-flash
CompareGet Started

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.

Context Window
1M
Max Output Tokens
384K
Released
2026-04-24
Capabilities
Function CallingPrompt Caching
Available Providers
DeepSeekDeepSeekAzureAzureBaiLianAliyun
Supported Protocols
OpenAIopenaiAnthropicanthropic

Providers

BaiLianAliyun
Input Tokens
$0.14/M
Output Tokens
$0.28/M
Cache Read
$0.028/M
Cache Write
$0.175/M
Web Search
$0.01/R
Protocols
Anthropicanthropic
AzureAzure
Input Tokens
$0.19/M
Output Tokens
$0.51/M
Cache Read
$0.19/M
Web Search
$0.035/R
Protocols
OpenAIopenai/v1/chat/completions/v1/responses
Anthropicanthropic
DeepSeekDeepSeek
Input Tokens
$0.14/M
Output Tokens
$0.28/M
Cache Read
$0.0028/M
Protocols
OpenAIopenai/v1/chat/completions
Anthropicanthropic

Code Examples

from openai import OpenAI
client = OpenAI(
base_url="https://api.ofox.ai/v1",
api_key="YOUR_OFOX_API_KEY",
)
response = client.chat.completions.create(
model="deepseek/deepseek-v4-flash",
messages=[
{"role": "user", "content": "Hello!"}
],
)
print(response.choices[0].message.content)

Uptime & Status

Benchmarks

LMArenaEvaluated as deepseek-v4-flash

DeepSeek V4 Flash scores 1438 in the Overall category of the LMArena text leaderboard (style control), ranking #70 of 374 models based on 41,568 human preference votes (updated 2026-07-12).

Benchmark scores for deepseek-v4-flash on LMArena
CategoryArena Score95% CIRankVotes
Overall14331442#70 of 37441,568
Hard Prompts14551466#65 of 37427,503
Coding14761489#71 of 36912,085
Math14141441#78 of 3622,133
Creative Writing14011418#65 of 3726,649
Instruction Following14231436#66 of 37413,952
Chinese14601489#73 of 3442,014

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.

Frequently Asked Questions

DeepSeek V4 Flash on Ofox.ai costs $0.14/M per million input tokens and $0.28/M per million output tokens. Pay-as-you-go, no monthly fees.