Gemini 3 Flash Preview
gemini-3-flash-previewGemini 3 Flash is Google's efficient model, strengthening general knowledge and instruction following while staying fast — suited to localization, customer support and social-listening at high volume.
Pricing
Tune and Pro share the same prices.
| Input context | Input | Output | Cache read | Cache write |
|---|---|---|---|---|
| - | $0.5 / 1M tokens | $3 / 1M tokens | $0.05 / 1M tokens | $0.5 / 1M tokens |
Capabilities
Gemini 3 Flash scored 74 overall, performing well — an efficient model that keeps quality at speed. It stands out in general knowledge (91%), instruction following (89%) and multilingual capability (88%), staying reliable and format-compliant across languages, which fits high-volume translation and support work. Comparatively, agent collaboration (63%) is weak; multi-step tool chains need simplification. Fast, rule-following multilingual responses are its strength.
Advanced Capabilities
8 items18 questions · Mixed
12 questions · Objective
6 questions · Mixed
22 questions · Subjective
9 questions · Mixed
16 questions · Mixed
14 questions · Mixed
11 questions · Subjective
General Foundations
3 items10 questions · Mixed
23 questions · Subjective
16 questions · Mixed
Safety & Compliance
2 items15 questions · Subjective
18 questions · Subjective
Performance
Gemini 3 Flash Preview
gemini-3-flash-preview- First token · TTFT · P50
- 978ms P95 1267 ms
- Output speed · TPS · P50
- 69tok/s P95 91 tok/s
- Full response · E2E · P50
- 6.10s P95 9.12 s
- Request success rate
- 97.66% 161,817 completed requests
Gemini 3 Flash Preview example code and API
gemini-3-flash-preview/api/v0/chat/completions/api/v1/chat/completionscurl https://silvamux.io/api/v1/chat/completions \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gemini-3-flash-preview",
"messages": [{"role": "user", "content": "Hello, please introduce yourself"}]
}'# pip install openai
from openai import OpenAI
client = OpenAI(
base_url="https://silvamux.io/api/v1",
api_key="YOUR_API_KEY",
)
response = client.chat.completions.create(
model="gemini-3-flash-preview",
messages=[
{"role": "user", "content": "Hello, please introduce yourself"}
],
)
print(response.choices[0].message.content)// Gradle: implementation(__HL_SLOT_0__)
import okhttp3.*;
OkHttpClient client = new OkHttpClient();
MediaType json = MediaType.get("application/json");
String payload = "{\"model\":\"gemini-3-flash-preview\",\"messages\":[{\"role\":\"user\",\"content\":\"Hello, please introduce yourself\"}]}";
Request request = new Request.Builder()
.url("https://silvamux.io/api/v1/chat/completions")
.header("Authorization", "Bearer YOUR_API_KEY")
.post(RequestBody.create(payload, json))
.build();
try (Response response = client.newCall(request).execute()) {
System.out.println(response.body().string());
}// npm install openai
import OpenAI from 'openai'
const client = new OpenAI({
baseURL: 'https://silvamux.io/api/v1',
apiKey: 'YOUR_API_KEY',
})
const response = await client.chat.completions.create({
model: 'gemini-3-flash-preview',
messages: [{ role: 'user', content: 'Hello, please introduce yourself' }],
})
console.log(response.choices[0].message.content)