Models

Available models

Use these model ids in MGPT Platform requests. Provider costs are shown per 1M tokens, separating cache-miss input from cheaper cache reads where available.

Other Models

Model	Type	Provider	Input miss / 1M	Cache read / 1M	Output / 1M
`deepseek-v4-pro`	Text	DeepSeek	`$0.435`	`$0.0036`	`$0.87`
`deepseek-v4-flash`	Text	DeepSeek	`$0.14`	`$0.0028`	`$0.28`
`gemini-3.1-flash-lite`	Text	Google	`$0.25`	N/A	`$1.50`
`gpt-oss-120b`	Text	OpenAI	`$0.039`	N/A	`$0.18`
`qwen3.6-27b`	Text	Qwen	`$0.2596`	N/A	`$2.385`
`qwen3.6-35b-a3b`	Text	Qwen	`$0.14`	N/A	`$1.00`
`qwen3.5-9b`	Text	Qwen	`$0.10`	N/A	`$0.15`
`minimax-m3`	Text	MiniMax	`$0.30`	`$0.060`	`$1.20`
`mimo-v2.5-pro`	Text	Xiaomi	`$0.435`	`$0.0036`	`$0.87`
`mimo-v2.5`	Text	Xiaomi	`$0.14`	`$0.0028`	`$0.28`

Continual MI Models

Model	Type	Provider	Input miss / 1M	Cache read / 1M	Output / 1M
No Continual MI models are currently exposed on MGPT Platform.