DzTokens

Gemini 2.5 Flash

VisionReasoningTool use

Google's fast and cost-effective model with a 1M token context window. Best for high-volume, low-latency tasks and agentic use cases.

Model id — paste this in your code
google/gemini-2.5-flash

Pricing · per 1M tokens

1 Pulse = 1 DA
Input54 Pulses= 54 DA
Output450 Pulses= 450 DA
Cached input5.4 Pulses= 5.4 DA
Context window1.05MMax output: 66K

Capabilities

  • Vision — Understands images and documents you send
  • Reasoning — Thinks step by step before answering
  • Tool use — Can call your functions / tools

Estimate your cost

Estimated cost

Use this model

Create an account, top up in dinars and call it with your key:

from openai import OpenAI

client = OpenAI(
    base_url="https://dztokens.com/v1",
    api_key="sk-dz-…",
)

reply = client.chat.completions.create(
    model="google/gemini-2.5-flash",
    messages=[{"role": "user", "content": "Salam!"}],
)
print(reply.choices[0].message.content)
Create a free account Console

Start building with AI today

Create your account, top up from 500 DA and send your first request in minutes.

Create a free account Console