Gemini 3.7 Flash

Gemini 3.7 Flash is Google's August 2026 refresh of the fast-tier Gemini, and it is a coding and agent model first. Against 3.6 Flash it lifts FrontierCode from 34.4% to 43.6%, DeepSWE from 48.6% to 65.3%, and Terminal-bench from 78.0% to 85.8%, while keeping the same 1M-token context and 64K output ceiling. Artificial Analysis clocked it the fastest of 186 models at roughly 340 tokens per second, and it exposes tunable thinking levels so you can trade latency for depth per call. Reach for it as a high-volume default for coding assistants, agent loops, and long-document work where throughput and cost per task both matter.

Quick start

Use any OpenAI-compatible client. Just change base_url and your key.

from openai import OpenAI

client = OpenAI(
    api_key="tl-xxxxxxxxxxxxxxxxxxxxxxxx",
    base_url="https://api.thalam.ai/v1",
)

response = client.chat.completions.create(
    model="google/gemini-3.7-flash",
    messages=[
        {"role": "user", "content": "Hello from Thalam!"}
    ],
)

print(response.choices[0].message.content)