超稳API / Stable API vs Modelflare

One model on two relays: integrity, speed, price, cache and availability, plus intelligence adjusted for degradation and value for money, with advice for the priority you choose.

Model
Alibaba-2
qwen-stable

Which to choose

Five dimensions

Further out is better. Integrity is the share of intelligence kept; the other four are places among all relays of this model.
IntegrityNot checkedSpeedNot checkedPriceCacheNot checkedAvailabilityNot checked
超稳API / Stable APIModelflare

What matters most to you

For Balanced, choose 超稳API / Stable API

Main reasons

Integrity25%

Integrity: no data for at least one relay

Availability20%

Availability: no data for at least one relay

Value20%

超稳API / Stable API gives 5.30 points per unit of money, 220% more than Modelflare at 1.66

Speed15%

Speed: no data for at least one relay

Price15%

超稳API / Stable API costs ¥6.36 per session, 69% less than Modelflare at ¥20.4

Cache5%

Cache: no data for at least one relay

超稳API / Stable API

Stronger
Price¥6.36Value5.30
Weaker
No clear gap

Modelflare

Stronger
No clear gap
Weaker
Price¥20.4Value1.66

Key figures

Adjusted intelligence

Higher is better

Cost per coding session

2M input + 40k output tokens · Lower is better

Value for money

Intelligence points per ¥1 · Higher is better

All figures

Intelligence

Adjusted intelligence33.733.7
Intelligence keptNot checked100%
Question accuracyNot checked94.4% (90 questions)

Cost

Value for money5.301.66
Cost per coding session¥6.36¥20.4
Input price¥3¥9.6
Output price¥9¥28.8
Cache discountNot providedNot provided

Performance

Time to first token—1.06 s
Availability, last 24 hours—98.8%
Samples—163

Route

GroupAlibaba-2qwen-stable
Data securityLevel 0Level 4
Adjusted intelligence = the model's published intelligence index × intelligence kept. Intelligence kept = the relay's question accuracy ÷ the highest accuracy among checked relays of the same model; under 10 questions, no adjustment. Value = adjusted intelligence ÷ cost per coding session. Gaps under a threshold count as close. Advice weighs the metrics by the chosen priority and is a guide only.