Models & pricing
17 open models behind one OpenAI-compatible endpoint (https://api.pgsgrove.com/v1). Prices are $ per 1M tokens, billed from prepaid credits — no subscription, no monthly minimum. Turbo ids route with priority. See the quickstart to make your first call.
| Model id | Modality | Context | Standard $/1M | Turbo $/1M |
|---|---|---|---|---|
mimo-v2.5MiMo V2.5 — Xiaomi |
Text + Image | 1M | $0.5 in · $2.5 out | $0.75 in · $3.75 out |
| Multimodal flagship: reasons over text and images in one long-context model. License: MIT. | ||||
mimo-v2.5-proMiMo V2.5 Pro — Xiaomi |
Text | 1M | $1.25 in · $3.75 out | $1.875 in · $5.625 out |
| The heavyweight MiMo — trillion-parameter reasoning at full context. License: MIT. | ||||
deepseek-v4-proDeepSeek V4 Pro — DeepSeek |
Text | 1M | $1.625 in · $3.25 out | $2.4375 in · $4.875 out |
| Frontier reasoning + coding at a 1M-token window. License: MIT. | ||||
deepseek-v4-flashDeepSeek V4 Flash — DeepSeek |
Text | 1M | $0.1125 in · $0.225 out | $0.1688 in · $0.3375 out |
| Fast, low-cost DeepSeek with the same 1M-token context. License: MIT. | ||||
glm-5.2GLM-5.2 — Z.ai |
Text | 1M | $1.1625 in · $3.75 out | $1.7438 in · $5.625 out |
| The newest GLM — million-token context and strong tool use. License: MIT. | ||||
glm-5.1GLM-5.1 — Z.ai |
Text | 198K | $1.3125 in · $4.375 out | — |
| Balanced GLM workhorse with prompt caching. License: MIT. | ||||
glm-5GLM-5 — Z.ai |
Text | 198K | $0.75 in · $2.6 out | — |
| Capable general-purpose GLM for everyday workloads. License: MIT. | ||||
glm-4.7-flashGLM-4.7 Flash — Z.ai |
Text | 198K | $0.075 in · $0.5 out | — |
| The value pick — tiny, quick, and remarkably cheap. License: MIT. | ||||
kimi-k2.7Kimi K2.7 — Moonshot AI |
Text + Image | 256K | $0.925 in · $4.375 out | $1.3875 in · $6.5625 out |
| Latest Kimi — trillion-parameter agentic coding with vision input. License: Modified MIT. | ||||
kimi-k2.6Kimi K2.6 — Moonshot AI |
Text + Image | 256K | $0.9375 in · $4.375 out | $1.4063 in · $6.5625 out |
| Proven Kimi generalist with strong tool use and vision. License: Modified MIT. | ||||
nemotron-3-ultraNemotron 3 Ultra — NVIDIA |
Text + Image | 256K | $0.625 in · $2.75 out | $0.9375 in · $4.125 out |
| NVIDIA’s flagship reasoning model, with image input. License: OpenMDW 1.1. | ||||
nemotron-3-superNemotron 3 Super — NVIDIA |
Text | 256K | $0.1063 in · $0.5 out | $0.1595 in · $0.75 out |
| Frontier-class reasoning quality at bargain-tier pricing. License: NVIDIA Open Model. | ||||
gemma-4-31bGemma 4 31B — Google |
Text + Image | 256K | $0.1625 in · $0.475 out | $0.2438 in · $0.7125 out |
| Efficient dense multimodal model — great price-to-capability. License: Gemma (Apache 2.0). | ||||
minimax-m3MiniMax M3 — MiniMax |
Text + Image | 512K | $0.375 in · $1.5 out | — |
| A highly competent coding model from MiniMax. (Built with MiniMax M3.) License: MiniMax Community. | ||||
minimax-m2.5MiniMax M2.5 — MiniMax |
Text | 192K | $0.1875 in · $1.4375 out | $0.2813 in · $2.1563 out |
| Lean sparse MoE tuned for fast, affordable throughput. License: Modified MIT. | ||||
kokoro-82mKokoro 82M — Hexgrad |
Text | tts | $0.775 / 1M characters | $0 in · $0 out |
| Natural, fast text to speech from a tiny open model. License: Apache-2.0. | ||||
embeddinggemma-300mEmbeddingGemma 300M — Google |
Text | embedding | $0.0025 in · $0 out | — |
| Compact 768-dimension embeddings for search and RAG. License: Gemma Terms. | ||||