google/gemini-3.5-flash-lite
Google Gemini 3.5 Flash-Lite — the cheapest, fastest model in the 3.5 line, 1M context.
Tune the request, see the current price ceiling, and inspect the real response without leaving this model page.
Sign in first to upload images to this vision model.
HTTP status 429 ("Too Many Requests") means that the user has sent too many requests in a given amount of time, essentially exceeding their allocated rate limit. When receiving this response, a well-behaved client should pause its requests and wait for the duration specified in the server's Retry-After header, if one is provided. To prevent repeatedly overloading the server, the client should also implement an exponential backoff strategy for subsequent retry attempts.
Gemini 3.5 Flash-Lite (gemini-3.5-flash-lite) went GA on July 21, 2026, alongside 3.6 Flash, and the model card says it is built on 3.1 Flash-Lite. Google positions it as a low-latency, low-cost model for subagents and high-volume automation: document parsing, structured extraction, classification and translation at scale.
The most direct difference from the Flash models of the same period is the default thinking level. 3.5 Flash-Lite defaults to minimal, trading depth for throughput, while 3.6 Flash defaults to medium. Google's advice is to leave minimal on for extraction, routing and classification, and to raise it to medium or high when the model acts as a subagent that calls tools, runs code or reasons over several steps; otherwise it tends to end tool use too early. Its API rules match 3.6 Flash: temperature, top_p and top_k are ignored, and requests ending on a prefilled model turn return a 400.
Compared with 3.1 Flash-Lite, Google's model card shows 54.2% vs 38.3% on SWE-Bench Pro and 74.0% vs 54.3% on OSWorld-Verified, and the knowledge cutoff moves to March 2026 (some domains still stop at January 2025). Google's deprecation table lists it as the recommended replacement for 3.1 Flash-Lite, and in September 2026 Google advised new projects to start on either 3.5 Flash-Lite or 3.8 Flash. Context is 1,048,576 tokens, with up to 65,536 output tokens.
RouteMux briefly forwarded it to Gemini 3.8 Flash from October 24, 2026; it now runs the original model again and is billed at its own price.
On RouteMux, input is $0.09 per 1M tokens, output $0.75 and cached input $0.009, against Google's $0.3, $2.5 and $0.03.
Prices verified 2026-09-09
Cache read $0.009/M · Cache write —/M
thinking_level to medium or high.top_p and top_k have no effect, and prefilled model turns return a 400.thinking_level to medium or high when the work involves tool calls, code execution or multi-step reasoning.Limited time10% extra Credits on your first top-up — up to 200 bonus Credits
Applies to your first top-up within 72 hours of signing up. Minimum $20, once per account.
Use any of these IDs to call this model via the API.
google/gemini-3.5-flash-liteReplace the ROUTEMUX_KEY placeholder with your API key. Create one →
from openai import OpenAI
client = OpenAI(
base_url="https://api.routemux.com/v1",
api_key="<ROUTEMUX_KEY>",
)
completion = client.chat.completions.create(
model="google/gemini-3.5-flash-lite",
messages=[{"role": "user", "content": "What is the meaning of life?"}],
)
print(completion.choices[0].message.content)Google Gemini 3.5 Flash-Lite — the cheapest, fastest model in the 3.5 line, 1M context.
Gemini 3.5 Flash-Lite supports a context window of up to 1M tokens, with up to 64K output tokens per request.
Through RouteMux, Gemini 3.5 Flash-Lite costs $0.09 for input and $0.75 for output (USD per 1M tokens), below the official list price.
Gemini 3.5 Flash-Lite is OpenAI-compatible: point your base URL at RouteMux and set the model field to google/gemini-3.5-flash-lite — no code changes needed.
Gemini 3.5 Flash-Lite accepts text, image, video, file as input and produces text as output.
Gemini 3.5 Flash-Lite was released on July 21, 2026.
23,066 requests · last 30 days