DeepSeek logo

Deepseek Flash

deepseek/deepseek-flash
Chat
DeepSeek|
Function CallingVisionReasoning

Deepseek Flash (deepseek/deepseek-flash) is an AI model from DeepSeek with a 1,000,000-token context window and 393,216 max output tokens, priced at $0.30/1M input and $1.20/1M output tokens. Available via the haimaker.ai OpenAI-compatible API.

Context Window
1M
tokens
Max Output
393K
tokens
Input Price
$0.30
/1M tokens
Output Price
$1.20
/1M tokens

Pricing schedule

This model’s rate changes by time of day. All times are shown in the schedule’s own timezone.

WindowHoursInput /1MOutput /1MCache read /1M
Off-peak00:00–09:00 Asia/Shanghai, 12:00–14:00 Asia/Shanghai, 18:00–24:00 Asia/Shanghai, 00:00–24:00 Asia/Shanghai$0.15$0.60$0.0030
Standard09:00–12:00 Asia/Shanghai, 14:00–18:00 Asia/Shanghai$0.30$1.20$0.0060

Overview

A large language model by DeepSeek with strong reasoning and coding capabilities.

Features & Capabilities

Modechat
Context Window1,000,000 tokens
Max Output393,216 tokens
Function CallingSupported
VisionSupported
ReasoningSupported
Web SearchNot supported
Url ContextNot supported

API Usage

from openai import OpenAI

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

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

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

Frequently Asked Questions

What is the context window of Deepseek Flash?

Deepseek Flash (deepseek/deepseek-flash) has a 1,000,000-token context window and supports up to 393,216 output tokens per request.

How much does Deepseek Flash cost?

Deepseek Flash is priced at $0.30 per 1M input tokens and $1.20 per 1M output tokens when accessed via the haimaker.ai OpenAI-compatible API. Rates vary by time of day: off-peak at $0.15 per 1M input tokens (00:00–09:00, 12:00–14:00, 18:00–24:00, 00:00–24:00 Asia/Shanghai) and standard at $0.30 per 1M input tokens (09:00–12:00, 14:00–18:00 Asia/Shanghai).

Does Deepseek Flash have off-peak pricing?

Yes. Deepseek Flash uses time-based pricing with off-peak at $0.15 per 1M input tokens (00:00–09:00, 12:00–14:00, 18:00–24:00, 00:00–24:00 Asia/Shanghai) and standard at $0.30 per 1M input tokens (09:00–12:00, 14:00–18:00 Asia/Shanghai). The rate applied to a request depends on when it is made.

What features does Deepseek Flash support?

Deepseek Flash supports function calling, vision, reasoning.

How do I use Deepseek Flash via API?

Send requests to https://api.haimaker.ai/v1/chat/completions with model "deepseek/deepseek-flash" using any OpenAI-compatible SDK. Authentication uses a Bearer API key from https://app.haimaker.ai.

Use Deepseek Flash with the haimaker API

OpenAI-compatible endpoint. Start building in minutes.

Get API Access

haimaker.ai 2026 - All Rights Reserved