Google has released Gemini 3.6 Flash, a production-ready model focused on coding, multimodal processing, and multi-step agent workflows.
The new model is available through the Gemini API using:
gemini-3.6-flash
What Changed?
Gemini 3.6 Flash improves on Gemini 3.5 Flash in several practical areas:
- 17% fewer output tokens on the Artificial Analysis Index
- Fewer reasoning steps, tool calls, and repeated execution loops
- Better code generation with fewer unwanted file changes
- Improved document, chart, image, audio, video, and PDF understanding
- Native support for function calling, code execution, search grounding, file search, and computer use
Google reports that the model scored 49% on DeepSWE, compared with 37% for Gemini 3.5 Flash. It also improved on OSWorld-Verified from 78.4% to 83%.
Context Window and Pricing
Gemini 3.6 Flash supports:
- 1,048,576 input tokens
- 65,536 output tokens
- $1.50 per million input tokens
- $7.50 per million output tokens
The input price remains unchanged from Gemini 3.5 Flash, while the output price drops from $9 to $7.50 per million tokens.
Basic Python Example
from google import genai
client = genai.Client()
response = client.interactions.create(
model="gemini-3.6-flash",
input="Review this codebase and create a safe migration plan."
)
print(response.output_text)
One Migration Warning
When moving to Gemini 3.6 Flash, remove deprecated parameters such as:
temperature
top_p
top_k
candidate_count
Applications using thinking_budget should migrate to the newer thinking_level setting.
Final Take
Gemini 3.6 Flash is not simply a faster chatbot model. Its real target is production agent systems that repeatedly inspect files, generate code, call tools, and complete multi-step tasks.
The lower output price and reduced token usage could make it a practical upgrade for teams already running Gemini-based coding agents or automation workflows.
Official resources: Gemini 3.6 Flash documentation and Google’s release announcement.
Top comments (0)