LLM API

Fast, cost-effective model for everyday tasks. Great balance of speed, intelligence, and cost for high-volume applications.

Anthropic's latest flagship Opus model, with a 1M-token context window by default and 128K max output tokens. Same pricing as Opus 4.5–4.8 ($5/$25 per M tokens) with prompt caching (read $0.50/M, write $6.25/M) and web search. Compatible with the Anthropic Messages and OpenAI Chat Completions formats.

Anthropic's most capable Opus-tier model, with a 1M-token context window (200K on some surfaces), 128K max output tokens, and knowledge cutoff to January 2026. Builds on Opus 4.7 with stronger long-horizon agentic coding, better tool triggering, and adaptive thinking that reasons only when a turn needs it. Compatible with the Anthropic Messages and OpenAI Chat Completions formats.

Anthropic's latest flagship model with reliable knowledge cutoff to January 2026 and 128K max output tokens. Builds on Opus 4.6 with improved reasoning, coding, and instruction-following while staying compatible with the Anthropic Messages and OpenAI Chat Completions formats.

Anthropic's most capable model. Delivers breakthrough performance in reasoning, coding, and complex analysis with enhanced safety and instruction following.

Anthropic's most powerful model for highly complex tasks. Exceptional at research, analysis, and creative projects requiring deep expertise.

Anthropic's newest Sonnet model, tuned for the best balance of speed, cost, and intelligence. Features a 1M-token context window, 64K max output tokens, and knowledge cutoff to August 2025. Supports adaptive thinking that reasons only when a turn needs it, vision, and tool use. Compatible with the Anthropic Messages and OpenAI Chat Completions formats.

The successor to Claude Sonnet 4.5. Anthropic's frontier Sonnet-tier model for complex reasoning, nuanced understanding, and coding tasks with exceptional instruction following, at the same token pricing as Sonnet 4.5.

Anthropic's most intelligent and capable Sonnet model. Best-in-class for complex reasoning, nuanced understanding, and coding tasks with exceptional instruction following.

DeepSeek's fast, low-cost V4 model (checkpoint V4-Flash-0731). Thinking mode is on by default, with a 1M-token context window and heavily discounted cache reads. Replaces the retired deepseek-chat / deepseek-reasoner.

DeepSeek's strongest V4 model (checkpoint V4-Pro-0813). Advanced reasoning with thinking mode on by default, a 1M-token context window, and discounted cache reads - still far cheaper than comparable frontier models.

Google's latest Flash-tier workhorse model. Flat pricing across all input modalities (text, image, video, audio, PDF) at $1.50/$9 per million tokens with no long-context premium, a 1M token context window, and 65,536 max output tokens. Cached inputs get a 90% discount.

Gemini 3.1 Flash with native image generation capabilities. Can generate images directly in chat responses alongside text. Features separate pricing for text and image output tokens.

Stable release of the ultra-lightweight Gemini 3.1 Flash Lite. The most cost-effective Gemini model at $0.25/$1.50 per million tokens, with cached input (90% discount), audio input, and batch (50% discount) support. Ideal for high-throughput, budget-conscious applications.

Ultra-lightweight variant of Gemini 3.1 Flash. The most cost-effective Gemini model with support for cached input and audio input. Ideal for high-throughput, budget-conscious applications.

Google's latest and most capable Gemini model in preview. Features dynamic pricing that adjusts based on context length, with enhanced pricing for inputs over 200K tokens.

Google's most advanced reasoning model with state-of-the-art multimodal understanding, PhD-level reasoning, and leading coding performance.

Google's premium image generation model within the Gemini 3 Pro family. Generates high-quality images directly in chat with the highest fidelity among Gemini image models. Image output tokens are priced at 10x text output tokens.

Google's most powerful Gemini model in preview. Features breakthrough reasoning, coding, and multimodal capabilities with the largest context window.

Google's fast and efficient model with built-in thinking capabilities. Great balance of speed, reasoning, and cost for high-volume applications.

The cheapest tier of the Gemini 2.5 family, optimized for high-volume, latency-sensitive workloads. Delivers 2.5-generation quality at a fraction of the cost, ideal for classification, extraction, and real-time chat at scale.

Google's most capable model with state-of-the-art reasoning and 1M token context. Excels at complex coding, math, and multi-document analysis.

Google's fastest and most capable model. Features a massive 1M token context window, native multimodal support, and real-time capabilities.

Ultra-lightweight version of Gemini 2.0 Flash optimized for maximum speed and minimal cost. Perfect for high-volume, latency-sensitive applications.

Google's text embedding model for generating vector representations. Optimized for semantic search, clustering, and similarity tasks.

Z.ai's GLM 5.2 served through the OpenRouter aggregator - a large-scale text reasoning model with a 1M-token context window, suited for long-horizon agent workflows and project-level software engineering. Billed at OpenRouter's actual usage cost.

Z.ai's GLM 4.7 served through the OpenRouter aggregator - a flagship text model with enhanced programming capabilities and more stable multi-step reasoning/execution, showing clear gains on complex agent tasks. 200K context, billed at OpenRouter's actual usage cost.

OpenAI's most capable model (GPT-6 generation, September 2026), built for the hardest end-to-end work: complex reasoning, coding, computer use, research and document creation. 1M token context window, 128K max output tokens, cached inputs at a 90% discount. Prompts above 272K input tokens are billed at 2x input / 1.5x output for the whole request.

The fast, low-cost tier of OpenAI's GPT-5.6 family (GA July 2026). Luna is built for high-throughput workloads at $0.20/$1.20 per million tokens, with a 1M token context window, 128K max output tokens, and cached inputs at a 90% discount. Prompts above 272K input tokens are billed at 2x input / 1.5x output for the whole request.

The flagship tier of OpenAI's GPT-5.6 family (GA July 2026). Sol delivers the strongest reasoning, coding, and multimodal performance of the generation with a 1M token context window, 128K max output tokens, and cached inputs at a 90% discount. Prompts above 272K input tokens are billed at 2x input / 1.5x output for the whole request.

The balanced tier of OpenAI's GPT-5.6 family (GA July 2026). Terra offers near-flagship quality at $2/$12 per million tokens (40% of Sol), with a 1M token context window, 128K max output tokens, and cached inputs at a 90% discount. Prompts above 272K input tokens are billed at 2x input / 1.5x output for the whole request.

OpenAI's newest flagship model with a 1.05M token context window and 128K max output tokens. Supports cached inputs at 10× discount and improved reasoning, coding, and multimodal performance over the GPT-5.4 series.

OpenAI's newest flagship model with 1M context window and 128K output tokens. Delivers top-tier reasoning across all domains with adjustable reasoning effort levels from none to xhigh.

Fast and cost-efficient variant of GPT-5.4 with 400K context window and 128K output tokens. Excellent balance of performance and affordability for everyday tasks.

Ultra-lightweight and fastest GPT-5.4 variant with 400K context and 128K output. Designed for high-throughput, low-latency applications at minimal cost. Supports MCP for tool integration.

Maximum-capability tier of the GPT-5.2 family (dated snapshot). Thinks longer with more compute to deliver the most reliable answers on the hardest reasoning and coding tasks.

OpenAI's latest and most advanced GPT model. Delivers state-of-the-art performance across reasoning, coding, and creative tasks with enhanced capabilities.

Latest Codex generation built on the GPT-5.2 base. Brings GPT-5.2's stronger reasoning to agentic coding harnesses, served via the OpenAI Responses API.

Dated snapshot of GPT-5.1 for reproducible results. Supports cached input tokens for cost savings on repeated context. Ideal for production deployments requiring model version pinning.

GPT-5.1 variant optimized for agentic coding in Codex-style harnesses. Tuned for long tool-use loops, code editing, and repository-scale tasks, and served via the OpenAI Responses API.

OpenAI's long-horizon agentic coding flagship in the GPT-5.1 Codex family. Supports compaction to keep very long coding sessions within context, and is served via the OpenAI Responses API.

Cheaper, faster member of the GPT-5.1 Codex family for lighter coding tasks. Keeps the agentic-coding tuning and Responses API interface at roughly one fifth of the Codex price.

OpenAI's latest flagship model. Delivers exceptional performance across reasoning, coding, and creative tasks with a massive 1M token context window and 32K output tokens. Supports vision, function calling, and JSON mode.

Fast and efficient variant of GPT-5. Delivers strong performance across reasoning, coding, and creative tasks with a 1M token context window and 32K output tokens, at a fraction of the cost of GPT-5.

Ultra-fast and lightweight variant of GPT-5. Designed for high-throughput, low-latency applications with a 1M token context window and 32K output tokens at minimal cost.

OpenAI's maximum-capability GPT-5 tier. Uses more compute to think longer and deliver the most reliable answers on the hardest reasoning, coding, and analysis tasks.

OpenAI's most capable model for coding and instruction following. Features a 1M token context window, 32K output tokens, and major improvements in coding, complex prompts, and long-context tasks. 20% cheaper than GPT-4o on output.

A significant leap in small model performance. Matches or exceeds GPT-4o in intelligence while reducing latency by nearly half and cost by 83%. Ideal balance of speed, quality, and affordability.

OpenAI's fastest and cheapest model. Optimized for classification, autocompletion, and low-latency tasks. Ultra-affordable at $0.10/1M input tokens.

OpenAI's flagship multimodal model. Industry-leading performance in reasoning, coding, and creative tasks with native vision capabilities and structured output support.

Cost-effective, fast model with strong performance. Best for high-volume tasks where speed and cost matter more than absolute capability.

Full-size multimodal model with native audio input/output via chat completions. Handles speech in and speech out with strong general intelligence behind it.

Lightweight multimodal model with native audio input/output capabilities. Optimized for voice-based interactions and audio processing tasks.

OpenAI's open-weight gpt-oss-120b, a 117B-parameter Mixture-of-Experts model (5.1B active per forward pass) built for high-reasoning, agentic and general-purpose production use - served through OpenRouter only; it is not available on the direct OpenAI API. 131K-token context, billed at OpenRouter's actual usage cost.

xAI's flagship Grok model. Strongest reasoning and general capability in the Grok family, with a 500K-token context window.

xAI's coding-focused Grok model with a 500K-token context window - same token rates as Grok 4.6 with cheaper cache reads.

Cost-efficient Grok model with a 1M-token context window - under half the price of the 4.5/4.6 tier.

xAI's Grok 4.3 served through the OpenRouter aggregator - 1M-token context, vision input, billed at OpenRouter's actual usage cost. Prefer the direct grok-4.3 route unless you need the OpenRouter path.

Grok 4.20 Multi-Agent beta (0309 snapshot) - deep-research style multi-agent orchestration with a 1M-token context window. Served through the Responses API only.

Grok 4.20 non-reasoning variant (0309 snapshot) - fast responses with a 1M-token context window at the same rates as the reasoning variant.

Grok 4.20 reasoning variant (0309 snapshot) - chain-of-thought reasoning with a 1M-token context window. Reasoning tokens are billed additively on top of completion tokens.

xAI's agentic-coding Grok model with a 256K-token context window - the lowest rates in the Grok family.

Moonshot AI's Kimi K3 served through the OpenRouter aggregator - a 2.8T-parameter open-weight multimodal reasoning model with a 1M-token context and image/video input, built for complex coding, knowledge work, and long-horizon agentic workflows. Billed at OpenRouter's actual usage cost.

Moonshot AI's Kimi K2.7 Code served through the OpenRouter aggregator - a coding-focused member of the Kimi K2 family built to complete end-to-end programming tasks reliably over long contexts, with a native multimodal mixture-of-experts architecture, 256K context, and image input. Billed at OpenRouter's actual usage cost.

Meta's Llama 3.3 70B instruction-tuned multilingual model served through OpenRouter - text in/text out with 131K-token context, a strong open-weight general-purpose chat model. This id replaces the retired direct llama-3.3-70b-versatile route. Billed at OpenRouter's actual usage cost.

Meta's Llama 3.1 8B instruction-tuned model served through OpenRouter - a fast, efficient small open-weight model with 131K-token context for high-volume, low-cost workloads. This id replaces the retired direct llama-3.1-8b-instant route. Billed at OpenRouter's actual usage cost.

MiniMax's M3 multimodal foundation model served through OpenRouter - text, image and video input, 1M-token context with up to 512K output tokens, suited for long-horizon agentic work and coding. Billed at OpenRouter's actual usage cost.

MiniMax's M2.7 served through OpenRouter - an agentic large language model built for autonomous, real-world productivity with 204K-token context and multi-agent tool use. This id replaces the retired direct MiniMax-M2.7 route. Billed at OpenRouter's actual usage cost.

MiniMax's flagship M2-series language model, served through an OpenAI-compatible API. Strong multilingual capability (notably Chinese and English) at a very low price point ($0.30/$1.20 per million tokens) with prompt cache reads at $0.06/M.

Mistral Large 3 (2512) served through the OpenRouter aggregator - Mistral's most capable model to date, a sparse mixture-of-experts with 41B active parameters (675B total) released under Apache 2.0, with 256K context and image/file input. Billed at OpenRouter's actual usage cost.

Meta's Muse Spark 1.3 over the direct Meta API - a multimodal reasoning model for agentic work that accepts text, images, video, audio and PDF documents and returns text, with a 1M-token context window and cached-input pricing. Answers are generated after reasoning tokens, so keep max_tokens at 2,048 or more.

The discounted Contributor tier of Meta's Muse Spark 1.3 - same model, about 1/12 of the standard price, on the condition that prompts and responses may be used by Meta to train its models. Do not send personal, confidential or customer data. Shares an upstream limit of 100 requests / 3M tokens per minute.

Meta's Muse Spark 1.1 served through the OpenRouter aggregator - a multimodal reasoning model built for agentic tasks that accepts text, images, video, audio, and PDF documents and returns text, with a 1M-token context. Billed at OpenRouter's actual usage cost.

Fast, cost-effective reasoning model optimized for coding and STEM tasks. Provides strong reasoning at a fraction of the cost of larger reasoning models.

OpenAI's frontier reasoning model. Uses extended chain-of-thought to solve complex problems in science, coding, and math with high accuracy at a competitive price.

Highest-effort version of o3. Thinks longer with more compute to deliver the most reliable answers on the hardest science, math, and coding problems.

Efficient reasoning model that delivers strong performance at lower cost. Ideal for tasks requiring reasoning without the overhead of larger models.

Legacy pro-tier reasoning model. Uses more compute than o1 for the most reliable answers on hard problems. Predecessor to o3-pro, which offers better value.

OpenAI's most advanced reasoning model. Uses extended thinking time to solve complex problems in science, coding, and math with exceptional accuracy.

Qwen3.8 27B served through the OpenRouter aggregator - an open-weight dense vision-language model for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking mode and a 256K-token context. Billed at OpenRouter's actual usage cost.

Qwen3.8 2.4T-A95B served through the OpenRouter aggregator - the open-weight sparse mixture-of-experts variant of Qwen3.8 Max, with 95B active parameters out of 2.4T total. Text-only reasoning model with a 1M-token context and up to 262K output tokens, billed at OpenRouter's actual usage cost.

Alibaba's Qwen3.8 Max served through the OpenRouter aggregator - the flagship of the Qwen3.8 series and GA successor to Qwen3.8 Max Preview. A multimodal reasoning model for complex reasoning, visual understanding, and agentic work with a 1M-token context, billed at OpenRouter's actual usage cost.

Alibaba's Qwen3.7 Flash served through the OpenRouter aggregator - a low-cost vision-language reasoning model for multimodal agents, visual coding, search, and computer interaction, with strengths in object recognition and spatial understanding. 1M-token context, billed at OpenRouter's actual usage cost.

Alibaba's Qwen3 Max served through the OpenRouter aggregator - an updated release built on the Qwen3 series with major improvements in reasoning, instruction following, multilingual support, and long-tail knowledge coverage. Text-only with a 256K-token context, billed at OpenRouter's actual usage cost.

Upstage's Solar Pro 4 served through the OpenRouter aggregator - a cost-efficient LLM with a 524K-token context window, built for long-horizon tasks and agentic workflows with strengths in office productivity and document-intensive work. Billed at OpenRouter's actual usage cost.

OpenAI's most capable text embedding model. Generates high-quality vector representations for semantic search, clustering, and similarity tasks where retrieval accuracy matters most.

OpenAI's cost-efficient text embedding model. Generates vector representations for semantic search, clustering, and similarity tasks at very low cost.