GPT-4o Omni
OPENAI:GPT-4O-OMNIOmni MultimodalOperationalLoad:64%SLA:99.99%Zero Queue
Native Multimodal Transformer (Audio/Vision/Text)
Base Rate
10 Credits / 1K Tokens
24h +15.20%
Flagship end-to-end multimodal model processing text, vision, and audio natively with unprecedented intelligence.
24h Total Calls
9.85Mcalls
+15.20% vs yesterday
Total Generations
7.42Minferences
+14.1% · Multimodal Alignment 96%
Avg Latency (P50)
220ms
TTFT 140ms · 118 tok/s
Availability SLA
99.99%
Failover <50ms
Model Telemetry & Volume Chart
OPENAI:GPT-4O-OMNIInvocations
430,955次+10.13%
Live Telemetry
Operational Performance
Production-grade Latency & Concurrency
P50 TTFT140 ms
P99 Latency460 ms
Throughput TPS
Peak Concurrency3,800 req
Packet Loss Rate0.0006%
Cost & Billing Matrix
Credits and Token Breakdown
Base Task Cost10 Credits
Input Cost / 1k Tokens¥0.015
Output Cost / 1k Tokens¥0.045
4K Upscale Multiplier1.0x
VIP Discount至高 25% 优惠
Technical Specifications
Parameters, Context & Resolutions
ArchitectureNative Multimodal Transformer (Audio/Vision/Text)
ParametersEstimated 200B+ MoE
Context Window128,000 tokens
Max ResolutionHigh-res Vision Tokens
Max DurationN/A
Global Edge Gateways
Live Region Latency & Load Balancing
East China (Shanghai)
25ms
South China (Guangzhou)
28ms
Hong Kong Gateway
32ms
Asia Pacific (Singapore)
48ms
US West (Silicon Valley)
124ms
EU Central (Frankfurt)
146ms