ant:reasoning-premiumThe deepest reasoning tier — thinking-capable models scoring 85 or above.
ant:reasoning-premium is the strictest pool available: models must support extended thinking and score at least 85, and the router weights quality at the maximum while accepting the highest latency and cost. Use it for the problems where nothing else has worked.
Live routing over the last 30 days
Requests
319
Success rate
3.13%
Avg latency
17ms
25 models across 10 providers currently qualify. Per-request limits (context size, cost caps, your model allowlist) narrow this further at routing time.
o1-pro
o1-pro
Claude Sonnet 5
o3
o3
o1
OpenAI: o3 Mini High
o1
GPT-5
OpenAI: o4 Mini High
OpenAI: o4 Mini (batch)
OpenAI: o3 Mini (batch)
Claude Sonnet 4
o4-mini
o4-mini
Qwen 3 235B
Sonar Deep Research
DeepSeek R1 671B (Venice)
Command A (03-2025)
MiniMax M2.7
Sonar Reasoning Pro
DeepSeek R1
o3-mini
Kimi K2.6
Showing the 24 highest-scoring of 25.
from openai import OpenAI
client = OpenAI(
base_url="https://api.antbase.ai/v1",
api_key="YOUR_ANT_API_KEY",
)
response = client.chat.completions.create(
model="ant:reasoning-premium",
messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)Every virtual pool accepts the same request shape — swap the model id to change the routing policy.
Browse all virtual models →