Pay only for what you actually use

Anthropic's most powerful Opus model. Best for complex reasoning, agentic tasks, and long-context work.

Anthropic's latest and most capable Opus model. Premium reasoning and agentic tasks.

Anthropic's most powerful model. Best for complex reasoning and agentic tasks.

Anthropic's latest Sonnet model. Balanced premium model for rich Roblox NPC and gameplay agents.

Anthropic Claude Opus 5.5 via Pollinations — flagship reasoning for demanding coding and long-context agentic work. Requires pro credits.

Balanced premium model for rich Roblox NPC and gameplay agents.

Quick premium responses with strong safety and instruction following.

Same Claude Haiku 4.5. Uses lite credits.

Google's most advanced Gemini 3.1 Pro — frontier reasoning and multimodal capabilities. Requires pro credits.

Google Gemini 3 Flash — fast and capable multimodal model.

Google Gemini 3.7 Flash — enhanced multimodal model with improved reasoning.

Google Gemini 3.5 Flash — capable multimodal model with reasoning.

Google Gemini 3.6 Flash — capable multimodal model with reasoning.

Google Gemini 3.8 Flash — enhanced multimodal model with improved reasoning.

Ultra-fast and cost-effective for high volume game traffic.

Google Gemini 3.1 Flash Lite — fast multimodal model.

Google Gemini 3.5 Flash Lite — fast multimodal model.

Google Gemma 4 31B instruction-tuned — powerful open model. Uses lite credits.

Google Gemma 4 26B MoE instruction-tuned — open model. Uses lite credits.

Google Gemma 4 26B MoE instruction-tuned — raw, unmodified output including reasoning tokens. Uses lite credits.

xAI Grok 4.2 Reasoning — deep reasoning model with vision. Requires pro credits.

xAI Grok 4.2 (non-reasoning) with vision. Uses lite credits.

xAI Grok 4.3 with vision and reasoning. Uses lite credits.

xAI Grok 4.5 with reasoning and tool calling. Uses lite credits.

xAI Grok 4.6 with vision, reasoning and tool calling. Uses lite credits.

OpenAI GPT-6 Astra — frontier reasoning for complex agentic, coding, and multimodal work. Uses lite credits.

OpenAI GPT-5.5 — frontier model with vision and reasoning. Requires pro credits.

OpenAI GPT-5.6 Sol — frontier model with vision and reasoning. Uses pro credits.

OpenAI GPT-5.6 Terra — frontier model with vision and reasoning. Requires pro credits.

OpenAI GPT-5.6 Luna — capable model with vision and reasoning. Uses lite credits.

OpenAI GPT-6 Sol — reasoning for complex coding and agentic workflows. Uses lite credits.

OpenAI GPT-6 Luna — efficient reasoning for focused, high-volume tasks. Uses lite credits.

OpenAI GPT-5 Chat Latest — chat-tuned frontier model with vision. Requires pro credits.

OpenAI GPT-5.4 — frontier model with vision and reasoning. Requires pro credits.

GPT-5.4 Mini — a capable mid-size model with vision support. Uses lite credits.

Strong default choice for smart assistants and game copilots.

Same GPT-5 Mini at a discounted special rate. Requires pro credits.

OpenAI GPT-4o Mini — fast multimodal model. Uses lite credits.

Fast lightweight model for responsive in-game conversations.

Same GPT-5 Nano at a discounted special rate. Requires pro credits.

GPT-5.4 Nano — fast and capable with vision support. Uses lite credits.

Same GPT-5.4 Nano at a discounted special rate. Requires pro credits.

OpenAI's open-source 120B model for powerful NPC dialogue and game logic.

OpenAI's open-source 20B model — lightweight NPC dialogue and game logic. Uses lite credits.

Agentic reasoning model for complex task orchestration.

Same DeepSeek V4 Flash at a discounted special rate. Requires pro credits.

DeepSeek V4 Pro. Uses lite credits.

MiniMax M3 — vision-capable reasoning model. Lite credits.

Nemotron Lightning 3.5 30B — fast MoE model. Lite credits.

Qwen 3.8 Max — large-scale flagship model with vision and reasoning. Lite credits.

Perplexity Sonar — live web search built in. Answers are grounded with up-to-date results from the internet. Requires pro credits.

Long-context agentic model for deep gameplay workflows.

Kimi K2.7 Code — vision and reasoning model optimised for coding tasks. Requires pro credits.

Moonshot Kimi K3.

Z.ai GLM 5.2 — long-context reasoning model. Requires pro credits.

Same GLM 5.2 at a discounted special rate. Requires pro credits.

Z.ai GLM 5.3 — better at complex coding and long-horizon tasks than GLM 5.2. Priced the same as GLM 5.2. Lite credits.

Z.ai GLM 5.3 Flash — faster, cheaper variant of GLM 5.3. Lite credits.

Muse Spark 1.3, contributor tier — vision and reasoning model. Note: this provider may use your prompts to help improve Meta's products. Uses lite credits.

NVIDIA's 550B parameter flagship model with tool calling. Requires pro credits.

NVIDIA's 120B parameter model for high-quality game AI at low cost.

NVIDIA's 550B parameter flagship model. Uses lite credits.

Google's diffusion-based Gemma 4 26B. Supports reasoning mode and tool calling. Uses lite credits.

NVIDIA Nemotron 3 nano omni 30B reasoning — free-tier model. Note: this provider may log your prompts and responses to train the model.

OpenAI's open 120B model, run for speed — same weights as GPT-OSS 120B. Uses lite credits.

OpenAI's open 20B model — the quickest route to a short answer. Uses lite credits.

Qwen 3.8 27B — vision, reasoning, and tool calling. Uses lite credits.

Qwen 3.6 27B — vision, reasoning, and tool calling. Uses lite credits.

Mimo v2.5 — multimodal model. Uses lite credits.

Mimo v2.5 Pro — multimodal model. Requires pro credits.

Alibaba Qwen 3.7 Plus — vision-capable reasoning model with tool use and strong multilingual performance. Uses lite credits.

Ezra — cool AI model. Uses lite credits.

Mercury 2 by Inception Labs — diffusion-based language model with reasoning and tool support. Uses lite credits.

Alibaba Qwen3Guard 8B safety classifier. Does not respond normally — returns a structured safety verdict: "Safety: Safe/Unsafe\nCategories: ..." instead of conversational text. Ideal for content moderation pipelines.

Strong and cheap agentic model designed with token efficiency in mind. Great for coding and agentic work. Uses lite credits.

Celeris 1 — a diffusion LLM tuned for very fast, short, structured responses like classification, extraction, and scoring. Uses lite credits.