Zashmodel 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
claude-prime
claude-award

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 checkedSpeedPriceCacheNot checkedAvailability
ZashmodelModelflare

What matters most to you

For Balanced, choose Zashmodel

Main reasons

Integrity25%

Integrity: no data for at least one relay

Availability20%

Zashmodel was available 100.0% of the last 24 hours; Modelflare 99.0%

Value20%

Zashmodel gives 14.4 points per unit of money, 493% more than Modelflare at 2.42

Speed15%

Modelflare answers in 1.21 s, 63% faster than Zashmodel at 3.30 s

Price15%

Zashmodel costs ¥3.08 per session, 83% less than Modelflare at ¥18.3

Cache5%

Cache: no data for at least one relay

Zashmodel

Stronger
Price¥3.08Availability100.0%Value14.4
Weaker
Speed3.30 s

Modelflare

Stronger
Speed1.21 s
Weaker
Price¥18.3Availability99.0%Value2.42

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 intelligence44.244.2
Intelligence keptNot checked100%
Question accuracyNot checked93.3% (90 questions)

Cost

Value for money14.42.42
Cost per coding session¥3.08¥18.3
Input price¥1.4¥8.3
Output price¥7¥41.5
Cache discountNot providedNot provided

Performance

Time to first token3.30 s1.21 s
Availability, last 24 hours100.0%99.0%
Samples8190

Route

Groupclaude-primeclaude-award
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.