Gemma 4 26B A4B
Google's efficient Gemma 4 variant, which uses only part of its parameters per token, so it costs less to run.
- Model ID
gemma-4-26b-a4b- Maker
- Google (United States)
- Context
- 262,000 tokens
- Max output
- 262,000 tokens
- Input
- €0.13 per million tokens
- Output
- €0.63 per million tokens
- Tools
- Yes
- JSON mode
- Yes
- Processing
- In the EU
Example
from openai import OpenAI
client = OpenAI(base_url="https://api.powerspring.eu/v1", api_key="sk-ps-...")
reply = client.chat.completions.create(
model="gemma-4-26b-a4b",
messages=[{"role": "user", "content": "Hello"}],
)
print(reply.choices[0].message.content)