Singapore-based GPU cloud offering serverless inference for a broad catalog of open-weight and frontier models. Hosts Qwen, MiniMax, DeepSeek, Llama, and others with OpenAI-compatible endpoints. Focuses on Asia-Pacific availability and competitive pricing. Inference platform · OpenAI-compatible API · Asia Pacific · Gpu Cloud · Low Cost · Open Source · Serverless

Intelligence vs Price

Best value among GMI Cloud models on this chart: GPT-5.2 · GPT-5.1 · Kimi K2 Thinking (and 2 more on the dashed frontier). Prices use each model's lowest available GMI Cloud price across regions. Hover any dot for full pricing, or click a creator in the legend to isolate.

Language Models
Intelligence
Blended Price, $
Log X

GMI Cloud models

17 models in Global, 17 with pricing
All Model Types
All Creators
US Dollar ($)
Per 1M tokens
Input/1M
to
Output/1M
to
Select
Model
Creator
Input Price, $
Output Price, $
Context
Max Output
Inference Providers
Intelligence
Coding
GPT-5.2OpenAI logoOpenAI1.7514.00410K128K#1N/A
Claude Opus 4.5Anthropic logoAnthropic5.0025.00410K64K#2N/A
GPT-5.1OpenAI logoOpenAI1.2510.00410K128K#3#2
Claude Sonnet 4.5Anthropic logoAnthropic3.0015.001.0M64K#4#1
GPT-5OpenAI logoOpenAI1.2510.00410K128K#5#4
Kimi K2 ThinkingMoonshot AI (Kimi) logoMoonshot AI (Kimi)0.81.20262K100K#6N/A
DeepSeek V3.2DeepSeek logoDeepSeek0.280.4164K147K#7#3
MiniMax M2.1MiniMax logoMiniMax0.31.201.0M131K#8N/A
Claude Opus 4Anthropic logoAnthropic15.0075.00410K32K#9N/A
Claude Sonnet 4Anthropic logoAnthropic3.0015.001.0M64K#10#5
DeepSeek V3 324DeepSeek logoDeepSeek0.280.88164K66K#11#7
Qwen3 VL 235B A22B InstructAlibaba logoAlibaba0.31.40262K129K#12N/A
GPT-4oOpenAI logoOpenAI2.5010.00131K16K#13#6
GPT-4o miniOpenAI logoOpenAI0.150.6131K16K#14#8
Gemini 3 Flash PreviewGoogle logoGoogle0.53.001.0M66KN/AN/A
Gemini 3 Pro PreviewGoogle logoGoogle2.0012.001.0M66KN/AN/A
GLM-4.7 FP8Zhipu AI logoZhipu AI0.42.00203K16KN/AN/A