Virtual poolant:reasoning-fast

ANT Reasoning Fast

Thinking-capable models, tuned for latency instead of depth.

Extended thinking: Required

How this pool routes

ant:reasoning-fast keeps the hard requirement for thinking support but drops the quality floor and weights speed high. It suits reasoning work in an interactive loop, where waiting on a premium model would break the experience.

Best For

Interactive agentsQuick reasoningTool-use loops

Where requests went

Live routing over the last 30 days

Requests

137

Success rate

48.91%

Avg latency

312ms

ant:reasoning-fastUnknown51.1% · 70

qwen/qwen3-next-80b-a3b-thinkingSemantic-cache43.1% · 59

openai/o3-miniOpenRouter5.8% · 8

Models in this pool

1,528 models across 23 providers currently qualify. Per-request limits (context size, cost caps, your model allowlist) narrow this further at routing time.

Showing the 24 highest-scoring of 1,528.

Quick Start

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-fast",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)

Routing weights

Speed80/100

Quality60/100

Cost60/100

Selection criteria

Extended thinkingrequiredRequired

Speed weight80/100

Quality weight60/100

Cost weight60/100

Model IDant:reasoning-fast

Other pools

Every virtual pool accepts the same request shape — swap the model id to change the routing policy.

Browse all virtual models →