# NinjaChat Developers

- [NinjaChat documentation](https://docs.ninjachat.ai/index.md): Build with the NinjaChat API, connect your agent over MCP, and route requests across leading AI models.
- [Quickstart](https://docs.ninjachat.ai/quickstart.md): From your first API key to your first response.
- [TypeScript and Python SDKs](https://docs.ninjachat.ai/sdks.md): Install the official NinjaChat SDKs for typed responses, streaming, media generation, routing, usage, and webhooks.
- [Authentication](https://docs.ninjachat.ai/authentication.md): One header on every request.
- [OpenAI compatibility](https://docs.ninjachat.ai/openai-compatibility.md): Use the official OpenAI SDKs with NinjaChat — just change the base URL and API key.
- [Anthropic compatibility](https://docs.ninjachat.ai/anthropic-compatibility.md): Send Messages API requests through NinjaChat.
- [Text generation](https://docs.ninjachat.ai/chat.md): Generate text, use vision and tools, and route across providers with the SDK or raw REST.
- [Streaming](https://docs.ninjachat.ai/streaming.md): Stream responses token-by-token over server-sent events.
- [Conversation state](https://docs.ninjachat.ai/sessions.md): Keep conversation history in your application and send the relevant context on each stateless API call.
- [Batch](https://docs.ninjachat.ai/batch.md): Run up to 20 independent chat requests in one call, held up front and settled to actual token usage.
- [Model compare](https://docs.ninjachat.ai/compare.md): Run one prompt against 2–8 models in parallel and get the results ranked by quality, speed, cost, or a balance of all three.
- [Cost estimation](https://docs.ninjachat.ai/estimate.md): Price a chat request before you send it — the same pricing engine and token estimator that bill real traffic, with no key and no charge.
- [Pipelines](https://docs.ninjachat.ai/pipelines.md): Chain up to five chat, image, and video steps into one durable async job — billed once, polled once.
- [Presets](https://docs.ninjachat.ai/presets.md): Save a model, routing policy, system prompt, and parameters behind a slug — then send only messages.
- [Go live](https://docs.ninjachat.ai/production-readiness.md): A compact production checklist for reliable, observable, and cost-bounded NinjaChat integrations.
- [Observability](https://docs.ninjachat.ai/observability.md): Connect request IDs, routing traces, usage, balance, health, and signed webhooks into one operating loop.
- [Error handling](https://docs.ninjachat.ai/error-handling.md): Every error code, what it means, and what to do about it.
- [Rate limits](https://docs.ninjachat.ai/rate-limits.md): Limits per endpoint, response headers, and retry patterns.
- [Account management API](https://docs.ninjachat.ai/management.md): Manage API keys, projects, webhooks and account usage programmatically.
- [Image generation](https://docs.ninjachat.ai/image-generation.md): Generate and edit images with GPT Image, Nano Banana, FLUX and more.
- [Video generation](https://docs.ninjachat.ai/video-generation.md): Veo, Gemini Omni Flash, Kling, and Seedance behind one endpoint. Submit, poll, get an MP4.
- [Web search](https://docs.ninjachat.ai/search.md): Search the web and get AI-synthesized answers with cited sources.
- [Speech generation](https://docs.ninjachat.ai/speech.md): Turn text into audio through the API.
- [Embeddings](https://docs.ninjachat.ai/embeddings.md): Create vectors for semantic search, recommendations and retrieval.
- [Reranking](https://docs.ninjachat.ai/reranking.md): Sort retrieved documents by relevance before generating an answer.
- [Model catalog API](https://docs.ninjachat.ai/models.md): Quick guide to picking the right model for your use case.
- [Pricing](https://docs.ninjachat.ai/pricing.md): Chat bills per token — exact $/MTok, no flat tiers. Images, video, and search stay flat per unit.
- [NinjaChat MCP](https://docs.ninjachat.ai/mcp/overview.md): Generate images, videos and voiceovers in Claude, ChatGPT, Codex and other MCP clients.
- [MCP quickstart](https://docs.ninjachat.ai/mcp/quickstart.md): Connect NinjaChat, check the connection for free, then make your first creation.
- [Client setup](https://docs.ninjachat.ai/mcp/clients.md): One endpoint, eight clients. Pick yours.
- [Workflows](https://docs.ninjachat.ai/mcp/recipes.md): Copy a brief for UGC ads, launch kits, narrated videos and longer agent-led productions.
- [Voiceovers](https://docs.ninjachat.ai/mcp/voiceovers.md): Create narration, ad reads and spoken translations as downloadable MP3 or WAV files.
- [Tool reference](https://docs.ninjachat.ai/mcp/tools.md): Image, video and voice tools, saved references, cost estimates and publishing.
- [Billing & limits](https://docs.ninjachat.ai/mcp/billing.md): Know the price before creating. Control spending and resume work safely.
- [Ninja Router](https://docs.ninjachat.ai/router/overview.md): Automatically select models and providers for cost, speed and quality.
- [Smart routing](https://docs.ninjachat.ai/smart-routing.md): Choose a model automatically, then select a provider for cost, speed and data policy.
- [Fallbacks](https://docs.ninjachat.ai/fallback-chains.md): Provide ordered model candidates and let NinjaChat fail over across models and provider rails.
- [Spend controls](https://docs.ninjachat.ai/budget-routing.md): Set a per-request maximum while preserving intelligent routing and fallbacks.
- [Quality gates](https://docs.ninjachat.ai/quality-scoring.md): Evaluate important outputs in your application and escalate to a stronger model when needed.
- [AI models](https://docs.ninjachat.ai/models-overview.md): Explore chat, image and video models from OpenAI, Anthropic, Google, Black Forest Labs and more.
- [GPT-5.6 Sol](https://docs.ninjachat.ai/api-reference/models/chat/gpt-5-6-sol.md): OpenAI's flagship GPT-5.6 model for the hardest reasoning and coding workloads.
- [GPT-5.6 Terra](https://docs.ninjachat.ai/api-reference/models/chat/gpt-5-6-terra.md): OpenAI's balanced GPT-5.6 model for coding, agents, and professional work.
- [GPT-5.6 Luna](https://docs.ninjachat.ai/api-reference/models/chat/gpt-5-6-luna.md): OpenAI's high-volume GPT-5.6 model with 1.05M context and full tool support.
- [GPT-5.5 Pro](https://docs.ninjachat.ai/api-reference/models/chat/gpt-5-5-pro.md): OpenAI's highest-accuracy GPT-5.5 variant for difficult reasoning, tool use, and long-context professional work.
- [Claude Fable 5](https://docs.ninjachat.ai/api-reference/models/chat/claude-fable-5.md): Anthropic's Fable 5 for long-running agents, deep research, and complex multimodal work with a 1M-token context.
- [Claude Opus 5](https://docs.ninjachat.ai/api-reference/models/chat/claude-opus-5.md): Anthropic's Opus 5 for complex agentic coding and enterprise knowledge work.
- [Claude Sonnet 5](https://docs.ninjachat.ai/api-reference/models/chat/claude-sonnet-5.md): Anthropic's frontier Sonnet for coding and agents, with a native 1M context window.
- [Claude Opus 4.8](https://docs.ninjachat.ai/api-reference/models/chat/claude-opus-4-8.md): Anthropic's Opus 4.8 for deep reasoning, coding, visual analysis, and long-horizon agents.
- [Gemini 3.7 Flash](https://docs.ninjachat.ai/api-reference/models/chat/gemini-3-7-flash.md): Google's Gemini 3.7 Flash for coding, multimodal reasoning, and high-throughput agents.
- [Gemini 3.1 Pro](https://docs.ninjachat.ai/api-reference/models/chat/gemini-3-1-pro.md): Google's Gemini 3.1 Pro — top-tier reasoning with a 1M-token context.
- [Grok 4.6](https://docs.ninjachat.ai/api-reference/models/chat/grok-4-6.md): xAI's frontier Grok model for coding, agents, and visual knowledge work.
- [DeepSeek V4 Pro](https://docs.ninjachat.ai/api-reference/models/chat/deepseek-v4-pro.md): DeepSeek's current V4 Pro release for frontier coding and complex reasoning with a million-token context.
- [GLM 5.2](https://docs.ninjachat.ai/api-reference/models/chat/glm-5-2.md): Z.ai's GLM 5.2 agentic model for engineering, multilingual coding, and long-horizon tasks.
- [Claude Fable 5.1](https://docs.ninjachat.ai/api-reference/models/chat/claude-fable-5-1.md): Anthropic's Mythos-class Claude Fable 5.1 for hours-long, multi-application agent work.
- [Claude Haiku 4.5](https://docs.ninjachat.ai/api-reference/models/chat/claude-haiku-4-5.md): Anthropic's Haiku 4.5 with vision, tools, extended thinking, and a 64K output ceiling.
- [Claude Opus 4.6](https://docs.ninjachat.ai/api-reference/models/chat/claude-opus-4-6.md): Anthropic's Opus 4.6 for deep analysis and complex coding.
- [Claude Opus 4.7](https://docs.ninjachat.ai/api-reference/models/chat/claude-opus-4-7.md): Anthropic's Opus 4.7 for complex analysis, coding, multimodal work, and tool use.
- [Claude Sonnet 4.5](https://docs.ninjachat.ai/api-reference/models/chat/claude-sonnet-4-5.md): Anthropic's Sonnet 4.5 — strong at code and analysis.
- [Claude Sonnet 4.6](https://docs.ninjachat.ai/api-reference/models/chat/claude-sonnet-4-6.md): Anthropic's Sonnet 4.6 — strong code and reasoning at mid-tier cost.
- [Seed 1.8](https://docs.ninjachat.ai/api-reference/models/chat/seed-1-8.md): ByteDance's long-context multimodal Seed 1.8 reasoning model with a 256K context window.
- [Seed 2.0 Code](https://docs.ninjachat.ai/api-reference/models/chat/seed-2-0-code.md): ByteDance's coding-specialized multimodal Seed 2.0 Code model for repository-scale engineering.
- [Seed 2.0 Mini](https://docs.ninjachat.ai/api-reference/models/chat/seed-2-0-mini.md): ByteDance's low-cost multimodal Seed 2.0 Mini reasoning model with a 256K context window.
- [Seed 2.0 Pro](https://docs.ninjachat.ai/api-reference/models/chat/seed-2-0-pro.md): ByteDance's full-capability multimodal Seed 2.0 Pro reasoning model for professional agent workloads.
- [Seed 2.1 Turbo](https://docs.ninjachat.ai/api-reference/models/chat/seed-2-1-turbo.md): ByteDance's current multimodal Seed 2.1 Turbo model for long-context agents and visual reasoning.
- [Command A](https://docs.ninjachat.ai/api-reference/models/chat/command-a.md): Cohere's high-capability enterprise model for agents, retrieval, tool use, and multilingual production workloads.
- [Command R](https://docs.ninjachat.ai/api-reference/models/chat/command-r.md): Cohere's balanced enterprise model for long-context retrieval, structured work, and multilingual tool use.
- [Command R7B](https://docs.ninjachat.ai/api-reference/models/chat/command-r7b.md): Cohere's compact Command R model for inexpensive, latency-sensitive agents, retrieval, and multilingual work.
- [North Mini Code 1.0](https://docs.ninjachat.ai/api-reference/models/chat/north-mini-code.md): Cohere's compact coding model with 256K context and streaming text responses. This hosted route does not support image input, tool calls, or structured output.
- [DeepSeek R1 0528](https://docs.ninjachat.ai/api-reference/models/chat/deepseek-r1-0528.md): DeepSeek's open R1 reasoning model (May 2025 update) for deep reasoning, mathematics, and research agents.
- [DeepSeek V3](https://docs.ninjachat.ai/api-reference/models/chat/deepseek-v3.md): DeepSeek's non-reasoning V3 model for coding, math, and budget tool agents.
- [DeepSeek V3 0324](https://docs.ninjachat.ai/api-reference/models/chat/deepseek-v3-0324.md): DeepSeek's stable March 2025 V3 snapshot for deterministic, non-reasoning workloads.
- [DeepSeek V3.1](https://docs.ninjachat.ai/api-reference/models/chat/deepseek-v3-1.md): DeepSeek's hybrid instruct model served in deterministic non-thinking mode for fast coding, JSON, and tools.
- [DeepSeek V3.2](https://docs.ninjachat.ai/api-reference/models/chat/deepseek-v3-2.md): DeepSeek's V3.2 general-purpose model for budget agents, coding, and reasoning.
- [DeepSeek V4 Flash](https://docs.ninjachat.ai/api-reference/models/chat/deepseek-v4-flash.md): DeepSeek's current V4 Flash release with a million-token context.
- [DeepSeek V4 Flash 0731](https://docs.ninjachat.ai/api-reference/models/chat/deepseek-v4-flash-0731.md): DeepSeek's V4 Flash (July 31 build): fast, low-cost reasoning with a 1M-token context.
- [DeepSeek V4 Flash Vision Exp](https://docs.ninjachat.ai/api-reference/models/chat/deepseek-v4-flash-vision-exp.md): Experimental vision-enabled DeepSeek V4 Flash for image understanding and text reasoning. This route supports streaming but not tools or strict JSON schemas.
- [DeepSeek V4 Pro 0813](https://docs.ninjachat.ai/api-reference/models/chat/deepseek-v4-pro-0813.md): DeepSeek's V4 Pro (August 13 build) for agentic coding and production reasoning with a 1M-token context.
- [Gemini 2.5 Flash](https://docs.ninjachat.ai/api-reference/models/chat/gemini-2-5-flash.md): Google's high-efficiency Gemini 2.5 Flash.
- [Gemini 2.5 Flash-Lite](https://docs.ninjachat.ai/api-reference/models/chat/gemini-2-5-flash-lite.md): Google's efficient million-context multimodal model for high-volume classification, extraction, and lightweight agents.
- [Gemini 2.5 Pro](https://docs.ninjachat.ai/api-reference/models/chat/gemini-2-5-pro.md): Google's Gemini 2.5 Pro for long documents and research with a 1M-token context.
- [Gemini 3 Flash](https://docs.ninjachat.ai/api-reference/models/chat/gemini-3-flash.md): Google's Gemini 3 Flash with a 1M-token context window.
- [Gemini 3 Pro](https://docs.ninjachat.ai/api-reference/models/chat/gemini-3-pro.md): Google's Gemini 3 Pro for reasoning, coding, and multimodal work.
- [Gemini 3.1 Flash Lite](https://docs.ninjachat.ai/api-reference/models/chat/gemini-3-1-flash-lite.md): Google's low-cost million-token Gemini 3.1 Flash Lite for high-volume multimodal work.
- [Gemini 3.5 Flash](https://docs.ninjachat.ai/api-reference/models/chat/gemini-3-5-flash.md): Google's Gemini 3.5 Flash for multimodal agents, million-token analysis, and fast tool use.
- [Gemini 3.5 Flash-Lite](https://docs.ninjachat.ai/api-reference/models/chat/gemini-3-5-flash-lite.md): Google's Gemini 3.5 Flash-Lite for high-volume, low-cost agentic and multimodal workloads.
- [Gemini 3.6 Flash](https://docs.ninjachat.ai/api-reference/models/chat/gemini-3-6-flash.md): Google's high-efficiency Gemini 3.6 Flash for multimodal coding, agentic workflows, and long-context analysis.
- [Gemini 3.8 Flash](https://docs.ninjachat.ai/api-reference/models/chat/gemini-3-8-flash.md): Google's Gemini 3.8 Flash for coding, multimodal reasoning, and high-throughput agents.
- [Gemma 3 12B](https://docs.ninjachat.ai/api-reference/models/chat/gemma-3-12b.md): Google's compact multimodal Gemma 3 model for inexpensive vision and tool workloads.
- [Gemma 3 27B](https://docs.ninjachat.ai/api-reference/models/chat/gemma-3-27b.md): Google's larger open Gemma 3 model with reliable tool and JSON support.
- [Gemma 3 4B](https://docs.ninjachat.ai/api-reference/models/chat/gemma-3-4b.md): Google's efficient open 4B multimodal model with tools and structured outputs.
- [Gemma 4 26B A4B](https://docs.ninjachat.ai/api-reference/models/chat/gemma-4-26b-a4b.md): Google's sparse Gemma 4 variant with 4B active parameters for efficient multimodal reasoning.
- [Gemma 4 31B](https://docs.ninjachat.ai/api-reference/models/chat/gemma-4-31b.md): Google's popular open multimodal model with native reasoning and function calling.
- [Gemma 4 31B Turbo](https://docs.ninjachat.ai/api-reference/models/chat/gemma-4-31b-turbo.md): Google's Gemma 4 31B in a low-latency Turbo serving tier with multimodal input and function calling.
- [Gemma 4 31B Ultra](https://docs.ninjachat.ai/api-reference/models/chat/gemma-4-31b-ultra.md): Google's Gemma 4 31B in a higher-accuracy Ultra serving tier for reasoning and tool-driven text workloads.
- [Gemma 4 E4B](https://docs.ninjachat.ai/api-reference/models/chat/gemma-4-e4b.md): Google's tiny Gemma 4 E4B model with long-context reasoning and reliable tool calling.
- [MythoMax 13B](https://docs.ninjachat.ai/api-reference/models/chat/mythomax-13b.md): Gryphe's established 13B creative model for compact roleplay and dialogue workloads.
- [Ling 3.0 Flash](https://docs.ninjachat.ai/api-reference/models/chat/ling-3-0-flash.md): InclusionAI's 124B/5.1B-active MoE optimized for token-efficient production agents and tool execution.
- [Ling 3.0 Flash Fin](https://docs.ninjachat.ai/api-reference/models/chat/ling-3-0-flash-fin.md): InclusionAI's finance-tuned Ling model for investment research, multi-step analysis, and long-horizon financial workflows.
- [LongCat 2.0](https://docs.ninjachat.ai/api-reference/models/chat/longcat-2-0.md): Meituan's native million-token sparse-MoE model for agentic coding and long-horizon tool use.
- [Llama 3.1 70B](https://docs.ninjachat.ai/api-reference/models/chat/llama-3-1-70b.md): Meta's widely adopted Llama 3.1 70B instruct model for general agents, RAG, and multilingual chat.
- [Llama 3.1 8B Instruct](https://docs.ninjachat.ai/api-reference/models/chat/llama-3-1-8b.md): Meta's compact long-context Llama 3.1 8B instruct model for extreme-volume workloads.
- [Llama 3.3 70B Instruct Turbo](https://docs.ninjachat.ai/api-reference/models/chat/llama-3-3-70b.md): Meta's multilingual Llama 3.3 70B instruct model in a low-latency Turbo serving tier.
- [Llama 4 Maverick](https://docs.ninjachat.ai/api-reference/models/chat/llama-4-maverick.md): Meta's 128-expert multimodal Llama 4 Maverick model with a full 1M-token context and tool use.
- [Llama 4 Scout](https://docs.ninjachat.ai/api-reference/models/chat/llama-4-scout.md): Meta's efficient multimodal Llama 4 Scout model with tool use and a 327K-token context.
- [Muse Glimmer 30B](https://docs.ninjachat.ai/api-reference/models/chat/muse-glimmer-30b.md): Meta's fast 30B Muse Glimmer multimodal model with vision and function calling.
- [Phi-4](https://docs.ninjachat.ai/api-reference/models/chat/phi-4.md): Microsoft's compact 14B reasoning model with reliable structured output.
- [MiniMax M2.5](https://docs.ninjachat.ai/api-reference/models/chat/minimax-m2-5.md): MiniMax's efficient M2.5 reasoning model for budget agents, coding, and multilingual tool use.
- [MiniMax M2.7](https://docs.ninjachat.ai/api-reference/models/chat/minimax-m2-7.md): MiniMax's efficient agent model for complex harnesses and multi-step productivity work.
- [MiniMax M2.7 Turbo](https://docs.ninjachat.ai/api-reference/models/chat/minimax-m2-7-turbo.md): The high-throughput MiniMax M2.7 variant for reasoning-heavy agents and coding workflows.
- [MiniMax M3](https://docs.ninjachat.ai/api-reference/models/chat/minimax-m3.md): MiniMax's low-cost open-weight frontier model with native multimodality and 512K context.
- [Codestral 2508](https://docs.ninjachat.ai/api-reference/models/chat/codestral-2508.md): Mistral's low-latency coding model for chat generation, function calling, and repository-scale context.
- [Ministral 3 14B](https://docs.ninjachat.ai/api-reference/models/chat/ministral-3-14b.md): Mistral's largest Ministral 3 model for capable budget agents, visual document work, and multilingual coding.
- [Ministral 3 3B](https://docs.ninjachat.ai/api-reference/models/chat/ministral-3-3b.md): Mistral's smallest Ministral 3 model for inexpensive multilingual, visual, and function-calling workloads.
- [Ministral 3 8B](https://docs.ninjachat.ai/api-reference/models/chat/ministral-3-8b.md): Mistral's balanced Ministral 3 model for fast visual understanding, coding, and tool use.
- [Mistral Large](https://docs.ninjachat.ai/api-reference/models/chat/mistral-large.md): Mistral AI's Mistral Large for multilingual tasks, coding, and general-purpose work.
- [Mistral Large 3](https://docs.ninjachat.ai/api-reference/models/chat/mistral-large-3.md): Mistral's open-weight 675B/41B-active general-purpose flagship with native vision and tool use.
- [Mistral Medium 3.5](https://docs.ninjachat.ai/api-reference/models/chat/mistral-medium-3-5.md): Mistral's frontier-class multimodal model optimized for agentic coding and professional work.
- [Mistral Nemo 12B](https://docs.ninjachat.ai/api-reference/models/chat/mistral-nemo-12b.md): Mistral's compact multilingual 12B instruct model for bulk summarization and classification.
- [Mistral Nemo 12B TEE](https://docs.ninjachat.ai/api-reference/models/chat/mistral-nemo-12b-tee.md): Mistral's compact multilingual 12B model served in a TEE-backed confidential-compute environment.
- [Mistral Small 24B](https://docs.ninjachat.ai/api-reference/models/chat/mistral-small-24b.md): Mistral's efficient 24B instruct model with JSON support for structured generation and multilingual drafting.
- [Mistral Small 3.2 24B](https://docs.ninjachat.ai/api-reference/models/chat/mistral-small-3-2-24b.md): Mistral's inexpensive Small 3.2 multimodal instruct model with tools and structured JSON.
- [Mistral Small 4](https://docs.ninjachat.ai/api-reference/models/chat/mistral-small-4.md): Mistral's efficient 119B/6.5B-active hybrid model unifying instruct, reasoning, and coding.
- [Kimi K2.5](https://docs.ninjachat.ai/api-reference/models/chat/kimi-k2-5.md): Moonshot AI's native multimodal Kimi K2.5 for advanced coding, visual agents, and long-context reasoning.
- [Kimi K2.6](https://docs.ninjachat.ai/api-reference/models/chat/kimi-k2-6.md): Moonshot's multimodal agentic model for long-horizon coding and autonomous execution.
- [Kimi K2.7 Code](https://docs.ninjachat.ai/api-reference/models/chat/kimi-k2-7-code.md): Moonshot's coding-specialized Kimi with faster, more token-efficient long-horizon execution.
- [Kimi K2.7 Code Fast](https://docs.ninjachat.ai/api-reference/models/chat/kimi-k2-7-code-fast.md): Moonshot's high-throughput Kimi K2.7 Code tier for fast long-context engineering and structured code workloads.
- [Kimi K3](https://docs.ninjachat.ai/api-reference/models/chat/kimi-k3.md): Moonshot's Kimi K3 frontier model with native vision and a 1M-token context.
- [Kimi K3 Fast](https://docs.ninjachat.ai/api-reference/models/chat/kimi-k3-fast.md): Moonshot's latency-tuned Kimi K3 tier with native vision, tools, structured output, and a 1M-token context.
- [Nex N2 Pro](https://docs.ninjachat.ai/api-reference/models/chat/nex-n2-pro.md): Nex AGI's frontier agentic model for coding, multimodal reasoning, and long-horizon execution.
- [Ninja 1](https://docs.ninjachat.ai/api-reference/models/chat/ninja-1.md): NinjaChat's stable default-model alias, currently backed by ByteDance Seed 1.8.
- [Hermes 3 405B](https://docs.ninjachat.ai/api-reference/models/chat/hermes-3-405b.md): Nous Research's 405B Llama 3.1 fine-tune for high-quality general assistance and controlled generation.
- [Hermes 3 70B](https://docs.ninjachat.ai/api-reference/models/chat/hermes-3-70b.md): Nous Research's controllable Llama 3.1 fine-tune for assistants, structured output, and long-form conversation.
- [Nemotron 3 Nano 30B A3B](https://docs.ninjachat.ai/api-reference/models/chat/nemotron-3-nano.md): NVIDIA's efficient hybrid MoE reasoning model with 3B active parameters.
- [Nemotron 3 Nano Omni](https://docs.ninjachat.ai/api-reference/models/chat/nemotron-3-nano-omni.md): NVIDIA's compact Omni model for image-aware reasoning and function-driven agent workloads.
- [Nemotron 3 Super](https://docs.ninjachat.ai/api-reference/models/chat/nemotron-3-super.md): NVIDIA's 120B/12B-active hybrid MoE for compute-efficient multi-agent reasoning and tool workflows.
- [Nemotron 3 Ultra](https://docs.ninjachat.ai/api-reference/models/chat/nemotron-3-ultra.md): NVIDIA's 550B/55B-active frontier reasoning model for demanding agent workloads.
- [Nemotron 3.5 Lightning](https://docs.ninjachat.ai/api-reference/models/chat/nemotron-3-5-lightning.md): NVIDIA's hybrid Mamba-Transformer reasoning model with 3B active parameters.
- [Nemotron Nano 12B v2 VL](https://docs.ninjachat.ai/api-reference/models/chat/nemotron-nano-12b-v2-vl.md): NVIDIA's compact multimodal reasoning model for cost-efficient visual agents and tool workflows.
- [Nemotron Nano 9B V2](https://docs.ninjachat.ai/api-reference/models/chat/nemotron-nano-9b-v2.md): NVIDIA's compact 9B controllable-reasoning model with fast tool calls.
- [GPT-4.1](https://docs.ninjachat.ai/api-reference/models/chat/gpt-4-1.md): OpenAI's high-intelligence non-reasoning model with a one-million-token context window, vision, and tool use.
- [GPT-4.1 Mini](https://docs.ninjachat.ai/api-reference/models/chat/gpt-4-1-mini.md): OpenAI's efficient GPT-4.1 model with million-token context, vision, structured output, and tool use.
- [GPT-4o Mini](https://docs.ninjachat.ai/api-reference/models/chat/gpt-4o-mini.md): OpenAI's compact multimodal model for inexpensive, high-throughput chat, vision, and structured workflows.
- [GPT-5](https://docs.ninjachat.ai/api-reference/models/chat/gpt-5.md): A backwards-compatible NinjaChat API alias that currently serves OpenAI GPT-5.2.
- [GPT-5 Nano](https://docs.ninjachat.ai/api-reference/models/chat/gpt-5-nano.md): OpenAI's compact GPT-5 Nano for high-throughput reasoning, vision, and agent workloads.
- [GPT-5.3 Codex](https://docs.ninjachat.ai/api-reference/models/chat/gpt-5-3-codex.md): OpenAI's coding-specialized reasoning model for complex software engineering and agentic tool use.
- [GPT-5.4](https://docs.ninjachat.ai/api-reference/models/chat/gpt-5-4.md): OpenAI's GPT-5.4 flagship for hard reasoning and code.
- [GPT-5.4 Mini](https://docs.ninjachat.ai/api-reference/models/chat/gpt-5-4-mini.md): OpenAI's compact GPT-5.4 model for fast multimodal agents and structured workloads.
- [GPT-5.4 Nano](https://docs.ninjachat.ai/api-reference/models/chat/gpt-5-4-nano.md): OpenAI's smallest GPT-5.4 model for routing, extraction, and lightweight agent loops.
- [GPT-5.4 Pro](https://docs.ninjachat.ai/api-reference/models/chat/gpt-5-4-pro.md): OpenAI's highest-effort GPT-5.4 variant for the most demanding tasks.
- [GPT-5.5](https://docs.ninjachat.ai/api-reference/models/chat/gpt-5-5.md): OpenAI's GPT-5.5 flagship with a million-token context, vision, and tool use.
- [GPT-6 Astra](https://docs.ninjachat.ai/api-reference/models/chat/gpt-6-astra.md): OpenAI's GPT-6 Astra for complex reasoning and tool-using workflows, with image input and 1.05M context.
- [GPT-OSS 120B](https://docs.ninjachat.ai/api-reference/models/chat/gpt-oss-120b.md): OpenAI's Apache-2.0 open-weight 120B reasoning model for high-volume tool use.
- [GPT-OSS 120B Turbo](https://docs.ninjachat.ai/api-reference/models/chat/gpt-oss-120b-turbo.md): OpenAI's GPT-OSS 120B in a faster Turbo serving tier for tool-driven reasoning and coding workloads.
- [GPT-OSS 120B Ultra](https://docs.ninjachat.ai/api-reference/models/chat/gpt-oss-120b-ultra.md): OpenAI's GPT-OSS 120B in a high-speed Ultra serving tier with tools, JSON, and streaming.
- [GPT-OSS 20B](https://docs.ninjachat.ai/api-reference/models/chat/gpt-oss-20b.md): OpenAI's compact 20B open-weight reasoning model for ultra-fast agents and structured outputs.
- [o3-mini](https://docs.ninjachat.ai/api-reference/models/chat/o3-mini.md): OpenAI's o3-mini reasoning model for math, logic, and step-by-step problems.
- [Laguna XS 2.1](https://docs.ninjachat.ai/api-reference/models/chat/laguna-xs-2-1.md): Poolside's compact, low-cost agentic coding model with a 256K context window.
- [Qwen 2.5 72B](https://docs.ninjachat.ai/api-reference/models/chat/qwen-2-5-72b.md): Qwen's capable 72B instruction model for multilingual agents, code, math, and structured output.
- [Qwen 3.5 122B A10B](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-5-122b-a10b.md): Qwen's 122B/10B-active multimodal MoE, balancing stronger quality with efficient inference.
- [Qwen 3.5 27B](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-5-27b.md): Qwen's dense 27B multimodal model for capable coding, reasoning, and visual tool workflows.
- [Qwen 3.5 35B A3B](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-5-35b-a3b.md): Qwen's efficient 35B MoE activates only 3B parameters per token while supporting tools and vision.
- [Qwen 3.5 397B A17B](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-5-397b-a17b.md): Qwen's 397B/17B-active multimodal flagship with broad multilingual, coding, and agentic capability.
- [Qwen 3.5 9B](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-5-9b.md): Qwen's compact 9B multimodal reasoning model with tools and a 262K-token context.
- [Qwen 3.5 Flash](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-5-flash.md): Qwen's low-latency vision-language MoE for million-token multimodal analysis, portable tool use, and production agents.
- [Qwen 3.6 27B](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-6-27b.md): Qwen's compact 27B multimodal reasoning model for fast agents, agentic coding, and visual reasoning.
- [Qwen 3.6 35B A3B](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-6-35b-a3b.md): Qwen's sparse 35B/3B-active multimodal model for efficient coding and agents.
- [Qwen 3.6 Flash](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-6-flash.md): Qwen's fast native vision-language model for million-token multimodal reasoning, coding, and production agents.
- [Qwen 3.6 Plus](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-6-plus.md): Qwen's native vision-language Plus model for million-token multimodal reasoning, coding, and structured agent work.
- [Qwen 3.7 Flash](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-7-flash.md): Qwen's low-cost million-token multimodal model for fast reasoning, coding, video understanding, tools, and structured output.
- [Qwen 3.7 Max](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-7-max.md): Qwen's proprietary Max model for long-context coding, analysis, and multilingual knowledge work.
- [Qwen 3.7 Plus](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-7-plus.md): Qwen's fast multimodal flagship for agent loops, coding, and tool use.
- [Qwen 3.8 2.4T A95B](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-8-2-4t.md): Qwen's open-weight 2.4T-parameter sparse-MoE flagship with 95B active parameters.
- [Qwen 3.8 27B](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-8-27b.md): Qwen's efficient 27B multimodal model for cost-efficient coding, long-context agents, and multilingual work.
- [Qwen 3.8 Flash](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-8-flash.md): Alibaba's Qwen 3.8 Flash multimodal reasoning model for coding, agents, and document analysis.
- [Qwen 3.8 Flash Next](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-8-flash-next.md): Qwen 3.8 Flash Next for multimodal reasoning, coding, and tool-using agents with 256K context.
- [Qwen 3.8 Max](https://docs.ninjachat.ai/api-reference/models/chat/qwen-3-8-max.md): Qwen's 2.4T-parameter sparse-MoE frontier model for autonomous long-horizon work.
- [Qwen2.5 Coder 3B Instruct](https://docs.ninjachat.ai/api-reference/models/chat/qwen2-5-coder-3b.md): Qwen's tiny coding specialist for latency-sensitive and extreme-volume API workloads.
- [Qwen2.5 Coder 7B Instruct](https://docs.ninjachat.ai/api-reference/models/chat/qwen2-5-coder-7b.md): Qwen's compact coding specialist for inexpensive generation and structured code workflows.
- [Qwen3 14B](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-14b.md): Qwen's compact hybrid reasoning model for economical structured workloads.
- [Qwen3 235B A22B Instruct](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-235b-instruct.md): Qwen's 235B-parameter instruction model combines 22B active parameters with low-cost structured tool use.
- [Qwen3 30B A3B](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-30b-a3b.md): Qwen's sparse 30B model with only 3B active parameters for fast reasoning and tools.
- [Qwen3 32B](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-32b.md): Qwen's efficient dense 32B model served in fast non-thinking mode with direct tool support.
- [Qwen3 4B Instruct 2507](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-4b-instruct-2507.md): Qwen's compact 4B instruct model with a 256K context window.
- [Qwen3 4B Thinking 2507](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-4b-thinking-2507.md): Qwen's compact 4B visible-reasoning model with a 256K context window.
- [Qwen3 Coder 480B Turbo](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-coder-480b-turbo.md): Qwen's 480B MoE agentic coding model in a low-latency Turbo serving tier.
- [Qwen3 Coder Next](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-coder-next.md): Qwen's 80B/3B-active coding MoE for long-horizon software agents, repository navigation, and tool execution.
- [Qwen3 Max](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-max.md): Qwen's flagship non-reasoning model for fast tool-driven production workloads.
- [Qwen3 Max Thinking](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-max-thinking.md): Qwen's flagship thinking model with adaptive tool use and extended test-time reasoning.
- [Qwen3 Next 80B A3B](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-next-80b-a3b.md): Qwen's efficient 80B instruct model with 3B active parameters for RAG and stable tool use.
- [Qwen3 VL 235B A22B Instruct](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-vl-235b-a22b.md): Qwen's large sparse vision-language flagship for high-accuracy OCR, visual reasoning, coding, and tool orchestration.
- [Qwen3 VL 30B A3B Instruct](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-vl-30b-a3b.md): Qwen's efficient 3B-active vision-language model for image-aware agents, OCR, coding, and structured tool use.
- [Qwen3 VL 32B Instruct](https://docs.ninjachat.ai/api-reference/models/chat/qwen3-vl-32b-instruct.md): Qwen's dense 32B vision-language model for high-precision document recognition, spatial understanding, and image-aware agents.
- [QwQ 32B](https://docs.ninjachat.ai/api-reference/models/chat/qwq-32b.md): Qwen's QwQ 32B reasoning-focused open model, strong at math.
- [Euryale 3.1 70B](https://docs.ninjachat.ai/api-reference/models/chat/euryale-70b.md): Sao10K's Llama 3.1 70B creative model tuned for immersive dialogue and roleplay.
- [Lunaris 8B Turbo](https://docs.ninjachat.ai/api-reference/models/chat/lunaris-8b.md): Sao10K's Lunaris 8B creative chat model for dialogue-heavy and roleplay workloads.
- [Step 3.7 Flash](https://docs.ninjachat.ai/api-reference/models/chat/step-3-7-flash.md): StepFun's efficient Step 3.7 Flash reasoning and coding model for fast, high-volume agents.
- [HY3](https://docs.ninjachat.ai/api-reference/models/chat/hy3.md): Tencent's HY3 reasoning and coding model with a 262K-token context.
- [HY4 Preview](https://docs.ninjachat.ai/api-reference/models/chat/hy4-preview.md): Tencent's HY4 Preview mixture-of-experts model for coding agents, tool use, and long-context work.
- [Inkling](https://docs.ninjachat.ai/api-reference/models/chat/inkling.md): Thinking Machines Lab's 975B multimodal open-weight generalist with 1M context.
- [Inkling Small](https://docs.ninjachat.ai/api-reference/models/chat/inkling-small.md): Thinking Machines Lab's efficient 276B/12B-active multimodal reasoner with a 524K context window.
- [Solar Pro 4](https://docs.ninjachat.ai/api-reference/models/chat/solar-pro4.md): Upstage's long-context agentic model for multi-step work, tools, and evidence-heavy documents.
- [Grok 4](https://docs.ninjachat.ai/api-reference/models/chat/grok-4.md): xAI's Grok 4 with strong reasoning and up-to-date knowledge.
- [Grok 4.20](https://docs.ninjachat.ai/api-reference/models/chat/grok-4-20.md): xAI's million-token multimodal Grok 4.20 reasoning model for long-context agents and tool orchestration.
- [Grok 4.20 Fast](https://docs.ninjachat.ai/api-reference/models/chat/grok-4-20-fast.md): The non-reasoning Grok 4.20 endpoint for fast million-token multimodal and agentic workloads.
- [Grok 4.3](https://docs.ninjachat.ai/api-reference/models/chat/grok-4-3.md): xAI's Grok 4.3 million-token reasoning model with vision, tools, structured output, and December 2025 knowledge.
- [Grok 4.5](https://docs.ninjachat.ai/api-reference/models/chat/grok-4-5.md): xAI's frontier Grok 4.5 model for coding, STEM, and complex agent workflows.
- [Grok Build 0.1](https://docs.ninjachat.ai/api-reference/models/chat/grok-build-0-1.md): xAI's cost-efficient coding model tuned for agentic engineering and tool use.
- [MiMo V2.5](https://docs.ninjachat.ai/api-reference/models/chat/mimo-v2-5.md): Xiaomi's efficient multimodal agent model for coding and tool-driven workflows.
- [MiMo V2.5 Pro](https://docs.ninjachat.ai/api-reference/models/chat/mimo-v2-5-pro.md): Xiaomi's higher-capability MiMo V2.5 tier for long-context coding and tool-driven agents.
- [GLM 4.6](https://docs.ninjachat.ai/api-reference/models/chat/glm-4-6.md): Z.ai's efficient bilingual reasoning and coding model with structured tool support.
- [GLM 4.7](https://docs.ninjachat.ai/api-reference/models/chat/glm-4-7.md): Z.ai's full GLM 4.7 reasoning model for coding and multi-step agent work.
- [GLM 4.7 Flash](https://docs.ninjachat.ai/api-reference/models/chat/glm-4-7-flash.md): Z.ai's extremely low-cost GLM reasoning model for fast coding and tool workflows.
- [GLM-5](https://docs.ninjachat.ai/api-reference/models/chat/glm-5.md): Z.ai's GLM-5 with strong bilingual (Chinese/English) capabilities.
- [GLM 5.1](https://docs.ninjachat.ai/api-reference/models/chat/glm-5-1.md): Z.ai's agentic GLM 5.1 model with structured output, reasoning, and function-calling support.
- [GLM 5.2 Fast](https://docs.ninjachat.ai/api-reference/models/chat/glm-5-2-fast.md): Z.ai's speed-optimized GLM 5.2 for real-time agentic engineering with a full 1M-token context.
- [GLM 5.3](https://docs.ninjachat.ai/api-reference/models/chat/glm-5-3.md): Z.ai's frontier long-horizon coding model with mandatory adaptive reasoning and 1M context.
- [GLM 5.3 Flash](https://docs.ninjachat.ai/api-reference/models/chat/glm-5-3-flash.md): Z.ai's efficient GLM 5.3 variant for high-volume coding, multimodal agents, and million-token work.
- [Bria 3.2](https://docs.ninjachat.ai/api-reference/models/image/bria-3-2.md): Bria's 3.2 licensed-data text-to-image model.
- [Bria FIBO](https://docs.ninjachat.ai/api-reference/models/image/bria-fibo.md): Bria's FIBO licensed-data text-to-image model.
- [FLUX.1 Dev](https://docs.ninjachat.ai/api-reference/models/image/flux-1-dev.md): Black Forest Labs' open-weight FLUX.1 Dev.
- [FLUX 1.1 Pro Ultra](https://docs.ninjachat.ai/api-reference/models/image/flux-1-pro-ultra.md): Black Forest Labs' FLUX 1.1 Pro Ultra for up to 4MP output.
- [FLUX 1.1 Pro](https://docs.ninjachat.ai/api-reference/models/image/flux-1-1-pro.md): Black Forest Labs' FLUX 1.1 Pro for high-quality text-to-image.
- [FLUX.2 Dev](https://docs.ninjachat.ai/api-reference/models/image/flux-2-dev.md): Black Forest Labs' open-weight FLUX.2 Dev for text-to-image with reference images.
- [FLUX.2 Flex](https://docs.ninjachat.ai/api-reference/models/image/flux-2-flex.md): Black Forest Labs' FLUX.2 Flex with reference-image control and tunable detail.
- [FLUX.2 Klein](https://docs.ninjachat.ai/api-reference/models/image/flux-2-klein.md): Black Forest Labs' fastest, lowest-cost FLUX.2 model.
- [FLUX.2 Klein 4B](https://docs.ninjachat.ai/api-reference/models/image/flux-2-klein-4b.md): Black Forest Labs' 4B FLUX.2 Klein for the cheapest drafts.
- [FLUX.2 Klein 9B](https://docs.ninjachat.ai/api-reference/models/image/flux-2-klein-9b.md): Black Forest Labs' 9B FLUX.2 Klein for fast, low-cost generation.
- [FLUX.2 Max](https://docs.ninjachat.ai/api-reference/models/image/flux-2-max.md): Black Forest Labs' highest-quality FLUX.2 model.
- [FLUX.2 Pro](https://docs.ninjachat.ai/api-reference/models/image/flux-2-pro.md): Black Forest Labs' balanced FLUX.2 model for production text-to-image.
- [FLUX.1 Kontext Dev](https://docs.ninjachat.ai/api-reference/models/image/flux-kontext-dev.md): Black Forest Labs' open-weight Kontext editor; a reference image is required.
- [FLUX.1 Kontext Max](https://docs.ninjachat.ai/api-reference/models/image/flux-kontext-max.md): Black Forest Labs' highest-fidelity Kontext editing model.
- [FLUX.1 Kontext Pro](https://docs.ninjachat.ai/api-reference/models/image/flux-kontext-pro.md): Black Forest Labs' context-aware image editing with text instructions.
- [GPT Image 2](https://docs.ninjachat.ai/api-reference/models/image/gpt-image-2.md): OpenAI's GPT Image 2 for instruction-following image generation and editing.
- [Grok Imagine](https://docs.ninjachat.ai/api-reference/models/image/grok-imagine.md): xAI's Grok Imagine text-to-image model.
- [Grok Imagine Image 2.0](https://docs.ninjachat.ai/api-reference/models/image/grok-imagine-image-2-0.md): xAI's second-generation Grok Imagine image model.
- [Grok Imagine Image Quality](https://docs.ninjachat.ai/api-reference/models/image/grok-imagine-image-quality.md): xAI's highest-fidelity Grok Imagine image tier.
- [HiDream L1 Fast](https://docs.ninjachat.ai/api-reference/models/image/hidream-l1-fast.md): Generate images from text with Pruna's optimized HiDream model.
- [Nano Banana](https://docs.ninjachat.ai/api-reference/models/image/nano-banana.md): Google's Gemini 2.5 Flash Image for fast conversational image generation and editing.
- [Nano Banana 2](https://docs.ninjachat.ai/api-reference/models/image/nano-banana-2.md): Google's Gemini 3.1 Flash Image with 4K output and strong text rendering.
- [Nano Banana 2 Lite](https://docs.ninjachat.ai/api-reference/models/image/nano-banana-2-lite.md): Google's lightest Nano Banana 2 tier for high-volume image generation.
- [Nano Banana Pro](https://docs.ninjachat.ai/api-reference/models/image/nano-banana-pro.md): Google's Gemini 3 Pro Image for complex, multi-element compositions at 4K.
- [Pruna P-Image](https://docs.ninjachat.ai/api-reference/models/image/pruna-p-image.md): Pruna AI's P-Image for very low-cost drafts.
- [Qwen Image 2512](https://docs.ninjachat.ai/api-reference/models/image/qwen-image-2512.md): Qwen's December 2025 Image release with reference-image support.
- [Qwen Image Edit 2511](https://docs.ninjachat.ai/api-reference/models/image/qwen-image-edit-2511.md): Qwen's November 2025 image editor; a reference image is required.
- [Qwen Image Edit Max](https://docs.ninjachat.ai/api-reference/models/image/qwen-image-edit-max.md): Qwen's highest-quality image editor; a reference image is required.
- [Qwen Image Edit Plus](https://docs.ninjachat.ai/api-reference/models/image/qwen-image-edit-plus.md): Edit and combine up to three reference images with a text prompt.
- [Qwen Image Max](https://docs.ninjachat.ai/api-reference/models/image/qwen-image-max.md): Qwen's highest-quality Image tier.
- [Recraft V3](https://docs.ninjachat.ai/api-reference/models/image/recraft-v3.md): Recraft V3 for brand-grade vector and raster design assets.
- [Runway Gen-4 Image Turbo](https://docs.ninjachat.ai/api-reference/models/image/runway-gen4-image-turbo.md): Runway's Gen-4 Image Turbo; requires one to three reference images.
- [SDXL Turbo](https://docs.ninjachat.ai/api-reference/models/image/sdxl-turbo.md): Stability AI's SDXL Turbo for near-free single-step drafts.
- [Seedream 4.5](https://docs.ninjachat.ai/api-reference/models/image/seedream.md): ByteDance's Seedream 4.5 for fast, high-fidelity text-to-image.
- [Seedream 4.5](https://docs.ninjachat.ai/api-reference/models/image/seedream-4-5.md): ByteDance's Seedream 4.5 text-to-image model.
- [Seedream 5.0 Lite](https://docs.ninjachat.ai/api-reference/models/image/seedream-5-lite.md): ByteDance's lower-cost Seedream 5.0 tier.
- [Seedream 5.0 Pro](https://docs.ninjachat.ai/api-reference/models/image/seedream-5-pro.md): ByteDance's flagship Seedream 5.0 text-to-image model.
- [Stable Diffusion 3.5 Large](https://docs.ninjachat.ai/api-reference/models/image/stable-diffusion-3-5-large.md): Stability AI's Stable Diffusion 3.5 Large.
- [Wan 2.6 Image](https://docs.ninjachat.ai/api-reference/models/image/wan-2-6-image.md): Alibaba's Wan 2.6 text-to-image model.
- [Wan 2.7 Image Edit](https://docs.ninjachat.ai/api-reference/models/image/wan-2-7-image-edit.md): Alibaba's Wan 2.7 image editor; a reference image is required.
- [Z-Image Turbo](https://docs.ninjachat.ai/api-reference/models/image/z-image-turbo.md): Alibaba Tongyi's Z-Image Turbo for very low-cost text-to-image drafts.
- [Gemini Omni Flash](https://docs.ninjachat.ai/api-reference/models/video/gemini-omni-flash.md): Google's Gemini Omni Flash for fast text- and image-to-video with native audio.
- [Grok Imagine Video](https://docs.ninjachat.ai/api-reference/models/video/grok-imagine-video.md): xAI's Grok Imagine text- and image-to-video model.
- [Grok Imagine Video 1.5](https://docs.ninjachat.ai/api-reference/models/video/grok-imagine-video-1-5.md): xAI's Grok Imagine Video 1.5.
- [Kling 2.6 Pro](https://docs.ninjachat.ai/api-reference/models/video/kling-video.md): Kuaishou's Kling 2.6 Pro for fluid text- and image-to-video with audio.
- [Runway Gen-4.5](https://docs.ninjachat.ai/api-reference/models/video/runway-gen4-5.md): Runway's Gen-4.5 text- and image-to-video model.
- [Seedance 2.0](https://docs.ninjachat.ai/api-reference/models/video/seedance-2.md): ByteDance's Seedance 2.0 references-to-video model with audio.
- [Seedance 2.5](https://docs.ninjachat.ai/api-reference/models/video/seedance-2-5.md): ByteDance's Seedance 2.5: clips up to 15 seconds from text, a first and last frame, or reference images, video and audio, with native audio.
- [Seedance 2.0 Mini](https://docs.ninjachat.ai/api-reference/models/video/seedance-2-mini.md): ByteDance's compact Seedance 2.0 with audio.
- [Seedance 2.0 Fast](https://docs.ninjachat.ai/api-reference/models/video/seedance-2-0-fast.md): ByteDance's speed tier of Seedance 2.0.
- [Seedance 1 Lite](https://docs.ninjachat.ai/api-reference/models/video/seedance-lite.md): ByteDance's low-cost first-generation Seedance.
- [Seedance 1 Pro](https://docs.ninjachat.ai/api-reference/models/video/seedance-pro.md): ByteDance's first-generation Seedance Pro at 1080p.
- [Veo 3.1](https://docs.ninjachat.ai/api-reference/models/video/veo-3-1.md): Google's Veo 3.1 for cinematic text- and image-to-video with native audio.
- [Veo 3.1 Fast](https://docs.ninjachat.ai/api-reference/models/video/veo-3-1-fast.md): Google's speed-optimized Veo 3.1.
- [Wan 2.2 T2V A14B](https://docs.ninjachat.ai/api-reference/models/video/wan-2-2-t2v-a14b.md): Alibaba's open-weight Wan 2.2 with 14B active parameters for fixed 5-second clips.
- [Create a stateless response](https://docs.ninjachat.ai/api-reference/create-a-stateless-response.md): OpenAI Responses-compatible text, vision, structured-output, and function-calling API. Set stream=true for typed lifecycle events.
- [Create a chat completion](https://docs.ninjachat.ai/api-reference/create-a-chat-completion.md): OpenAI-compatible chat completions with ordered model fallbacks and provider routing policy. Pass model "@slug" to run a saved preset you own — the preset supplies models, prompt, tools, and routing; send only messages.
- [Run a saved preset](https://docs.ninjachat.ai/api-reference/run-a-saved-preset.md): Chat completion using a preset's saved model, routing, system prompt, and parameters. Send only your messages; pass model/models to override the preset's chain for this call. Response headers include X-Ninja-Preset and X-Ninja-Preset-Version.
- [Create an Anthropic-compatible message](https://docs.ninjachat.ai/api-reference/create-an-anthropic-compatible-message.md): Messages compatibility for text, image inputs, JSON output schemas, function tools and SSE. Accepts Authorization bearer or x-api-key. Uses the same metered gateway, routing and idempotency as Chat Completions. Native server tools, file IDs and signed thinking history are not supported; unsupported…
- [List live models](https://docs.ninjachat.ai/api-reference/list-live-models.md)
- [Get a live model](https://docs.ninjachat.ai/api-reference/get-a-live-model.md)
- [Get the live price sheet](https://docs.ninjachat.ai/api-reference/get-the-live-price-sheet.md): Public prices generated from the same tables used by billing.
- [Get public network statistics](https://docs.ninjachat.ai/api-reference/get-public-network-statistics.md)
- [Get this OpenAPI document](https://docs.ninjachat.ai/api-reference/get-this-openapi-document.md)
- [Generate images](https://docs.ninjachat.ai/api-reference/generate-images.md): storage is optional and defaults to durable, which copies output to NinjaChat storage and returns a permanent URL. Set storage=provider to skip that copy for lower latency; provider mode returns either a temporary provider URL or b64_json plus mime_type.
- [Create a video job](https://docs.ninjachat.ai/api-reference/create-a-video-job.md): Submit an async video job. Poll GET /videos/{id}, or register a signed webhook (POST /webhooks or https://www.ninjachat.ai/developers/webhooks) for video.completed / video.failed.
- [Get a video job](https://docs.ninjachat.ai/api-reference/get-a-video-job.md): Poll job status until completed or failed. Prefer a webhook if you do not want to poll.
- [Search the web](https://docs.ninjachat.ai/api-reference/search-the-web.md)
- [Generate speech](https://docs.ninjachat.ai/api-reference/generate-speech.md)
- [Create embeddings](https://docs.ninjachat.ai/api-reference/create-embeddings.md): Exact-model text embeddings. OpenAI text-embedding-3-small: $0.02 per million input tokens. Voyage voyage-4-large: $0.12 per million. Charges round up to the wallet's $0.0001 precision. No cross-model fallback or input truncation by default.
- [Rerank documents](https://docs.ninjachat.ai/api-reference/rerank-documents.md)
- [Run a batch of chat requests](https://docs.ninjachat.ai/api-reference/run-a-batch-of-chat-requests.md)
- [Compare multiple chat models](https://docs.ninjachat.ai/api-reference/compare-multiple-chat-models.md)
- [Estimate request cost](https://docs.ninjachat.ai/api-reference/estimate-request-cost.md): Public pre-flight estimate; does not deduct credits.
- [Create a multimodal pipeline](https://docs.ninjachat.ai/api-reference/create-a-multimodal-pipeline.md): Run up to five dependent chat, image and video steps as one durable async job.
- [Get a multimodal pipeline](https://docs.ninjachat.ai/api-reference/get-a-multimodal-pipeline.md)
- [Get model battle rankings](https://docs.ninjachat.ai/api-reference/get-model-battle-rankings.md)
- [Record an authenticated model ranking](https://docs.ninjachat.ai/api-reference/record-an-authenticated-model-ranking.md)
- [Get usage](https://docs.ninjachat.ai/api-reference/get-usage.md)
- [Get balance](https://docs.ninjachat.ai/api-reference/get-balance.md)
- [Get a request trace](https://docs.ninjachat.ai/api-reference/get-a-request-trace.md)
- [Get gateway health](https://docs.ninjachat.ai/api-reference/get-gateway-health.md)
- [List webhooks](https://docs.ninjachat.ai/api-reference/list-webhooks.md): List HTTPS endpoints registered for video completion and spend-alert callbacks. Console: https://www.ninjachat.ai/developers/webhooks
- [Create a webhook](https://docs.ninjachat.ai/api-reference/create-a-webhook.md): Register an HTTPS URL. Generic receivers get signed JSON; Discord and Slack incoming webhook URLs are detected automatically and receive rich alerts without NinjaChat signing-secret setup. Platform URL credentials are redacted in responses. New events require explicit subscription; omitted events re…
- [Delete a webhook](https://docs.ninjachat.ai/api-reference/delete-a-webhook.md)
- [List webhook deliveries](https://docs.ninjachat.ai/api-reference/list-webhook-deliveries.md): Recent delivery attempts for one endpoint, newest first.
- [Send a test webhook](https://docs.ninjachat.ai/api-reference/send-a-test-webhook.md): Send one synthetic event immediately to check your Discord/Slack channel or generic receiver signature handling before relying on live jobs.
- [management.webhooks.update](https://docs.ninjachat.ai/api-reference/managementwebhooksupdate.md): Requires webhooks:write. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 12…
- [management.webhooks.list](https://docs.ninjachat.ai/api-reference/managementwebhookslist.md): Requires webhooks:read. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120…
- [management.webhooks.create](https://docs.ninjachat.ai/api-reference/managementwebhookscreate.md): Requires webhooks:write. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 12…
- [management.whoami](https://docs.ninjachat.ai/api-reference/managementwhoami.md): Requires any management key. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits…
- [management.balance](https://docs.ninjachat.ai/api-reference/managementbalance.md): Requires usage:read. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120 re…
- [management.usage](https://docs.ninjachat.ai/api-reference/managementusage.md): Requires usage:read. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120 re…
- [management.audit](https://docs.ninjachat.ai/api-reference/managementaudit.md): Requires usage:read. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120 re…
- [management.keys.list](https://docs.ninjachat.ai/api-reference/managementkeyslist.md): Requires keys:read. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120 rea…
- [management.keys.create](https://docs.ninjachat.ai/api-reference/managementkeyscreate.md): Requires keys:write. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120 re…
- [management.keys.retrieve](https://docs.ninjachat.ai/api-reference/managementkeysretrieve.md): Requires keys:read. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120 rea…
- [management.keys.update](https://docs.ninjachat.ai/api-reference/managementkeysupdate.md): Requires keys:write. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120 re…
- [management.keys.revoke](https://docs.ninjachat.ai/api-reference/managementkeysrevoke.md): Requires keys:write. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120 re…
- [management.keys.rotate](https://docs.ninjachat.ai/api-reference/managementkeysrotate.md): Requires keys:write. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120 re…
- [management.projects.list](https://docs.ninjachat.ai/api-reference/managementprojectslist.md): Requires projects:read. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120…
- [management.projects.create](https://docs.ninjachat.ai/api-reference/managementprojectscreate.md): Requires projects:write. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 12…
- [management.projects.retrieve](https://docs.ninjachat.ai/api-reference/managementprojectsretrieve.md): Requires projects:read. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120…
- [management.projects.update](https://docs.ninjachat.ai/api-reference/managementprojectsupdate.md): Requires projects:write. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 12…
- [management.projects.archive](https://docs.ninjachat.ai/api-reference/managementprojectsarchive.md): Requires projects:write. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 12…
- [management.webhooks.delete](https://docs.ninjachat.ai/api-reference/managementwebhooksdelete.md): Requires webhooks:write. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 12…
- [management.webhooks.deliveries](https://docs.ninjachat.ai/api-reference/managementwebhooksdeliveries.md): Requires webhooks:read. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 120…
- [management.webhooks.test](https://docs.ninjachat.ai/api-reference/managementwebhookstest.md): Requires webhooks:write. Changing routing also requires routing:write. Budgets are integer USD cents. Responses are not cacheable. Keys/secrets are shown only at creation/rotation. Mutations do not support automatic retries: read current resource state after an ambiguous failure. Standard limits: 12…
- [Create a chat completion (legacy path)](https://docs.ninjachat.ai/api-reference/create-a-chat-completion-legacy-path.md): Supported compatibility surface — kept working indefinitely, but not the path for new integrations. Modern equivalent: `POST /responses`.
- [Generate images (legacy path)](https://docs.ninjachat.ai/api-reference/generate-images-legacy-path.md): Supported compatibility surface — kept working indefinitely, but not the path for new integrations. Modern equivalent: `POST /images/generations`.
- [Create a video job (legacy path)](https://docs.ninjachat.ai/api-reference/create-a-video-job-legacy-path.md): Supported compatibility surface — kept working indefinitely, but not the path for new integrations. Modern equivalent: `POST /videos`.
- [Get a video job (legacy path)](https://docs.ninjachat.ai/api-reference/get-a-video-job-legacy-path.md): Supported compatibility surface — kept working indefinitely, but not the path for new integrations. Modern equivalent: `GET /videos/{id}`.
- [Create a chat session](https://docs.ninjachat.ai/api-reference/create-a-chat-session.md): Supported compatibility surface — kept working indefinitely, but not the path for new integrations. Modern equivalent: `client-side message history on POST /responses`.
- [Get a chat session](https://docs.ninjachat.ai/api-reference/get-a-chat-session.md): Supported compatibility surface — kept working indefinitely, but not the path for new integrations. Modern equivalent: `client-side message history on POST /responses`.
- [Delete a chat session](https://docs.ninjachat.ai/api-reference/delete-a-chat-session.md): Supported compatibility surface — kept working indefinitely, but not the path for new integrations. Modern equivalent: `client-side message history on POST /responses`.
- [Export a chat session](https://docs.ninjachat.ai/api-reference/export-a-chat-session.md): Supported compatibility surface — kept working indefinitely, but not the path for new integrations. Modern equivalent: `client-side message history on POST /responses`.

## OpenAPI Specs

- [openapi](/openapi.json)

## Optional

- [Dashboard](https://www.ninjachat.ai/developers)
- [Playground](https://www.ninjachat.ai/developers/playground)
- [Changelog](https://www.ninjachat.ai/changelog)