Price over 30 days
New capacity entered the US West market.
MARKET INTELLIGENCE
See the market clearly.
Price discovery, supply signals, and execution quality across the inference economy.
Capacity utilization
Demand is approaching available supply.
Regional spot premium
Low-latency capacity is trading at a premium.
$13.92 inference prices
Model market share
24H notional volumeShare of $185.3M total daily market volume.
One market. Global reach.
11 ACTIVE REGIONSInference forward curve
Claude Sonnet · OutputSpot$13.92
Oct 2026$13.52
Nov 2026$12.95
Dec 2026$12.42
Q1 2027$11.80
Lock future inference capacity at transparent market prices. Fixed model, region, and service level.
Execution quality
24H99.999%Routing availability
82 msRouting overhead
8Supply integrations
11Active regions
QUALITY-ADJUSTED TOKENS
Optimize value, not just token cost.
| Model | Quality / 100 | Output / 1M | Latency | Price / quality point |
|---|---|---|---|---|
Claude SonnetAnthropic · US West | 94 | $13.92 | 182 ms | $0.148 |
GPT-5.6 SolOpenAI · US East | 97 | $10.48 | 164 ms | $0.108 |
Gemini ProGoogle · US West | 95 | $9.84 | 146 ms | $0.104 |
DeepSeek V4DeepSeek · US East | 89 | $1.62 | 218 ms | $0.018 |
Llama 4Meta · US West | 83 | $0.78 | 94 ms | $0.009 |
Qwen MaxAlibaba · Singapore | 90 | $4.12 | 156 ms | $0.046 |
Claude OpusAnthropic · US East | 98 | $23.76 | 242 ms | $0.242 |
Raw tokens are not equivalent across models. Illustrative workload scores are normalized to 100; price / quality divides output price by score. Compare within the same workload.
Prices, benchmarks, and infrastructure metrics are illustrative. They are not independently verified market observations.