Every model, one API key.

TypeSafe's System One decision model: send app state plus typed questions, get typed answers with calibrated probabilities in under a second.

$0.047 per 1M input tokens

Anthropic's flagship for complex agentic coding and long-horizon work, succeeding Opus 5.

$4.43 / $22.15 per 1M tokens

OpenAI's GPT-6 flagship for frontier reasoning, coding, and long-horizon agents.

$11.08 / $55.39 per 1M tokens

OpenAI's extended-thinking GPT-6 tier for the hardest reasoning and agent work.

$11.08 / $55.39 per 1M tokens

Anthropic's newest model for long-running agents and sustained reasoning.

$11.08 / $55.39 per 1M tokens

Anthropic's flagship for complex agentic coding and enterprise workloads.

$5.54 / $27.69 per 1M tokens

OpenAI's cost-efficient high-end GPT-6 model for demanding professional and agent work.

$2.22 / $11.08 per 1M tokens

GPT-6 Sol with pro-mode reasoning for higher-quality answers on complex tasks.

$2.22 / $11.08 per 1M tokens

Anthropic next-generation model for long-running agents.

$11.08 / $55.39 per 1M tokens

xAI's latest flagship for coding, knowledge work, and STEM tasks.

$2.22 / $6.65 per 1M tokens

OpenAI's latest flagship for complex reasoning, coding, and long-horizon agents.

$2.22 / $11.08 per 1M tokens

Z.ai's latest long-context reasoning model for complex agent and coding tasks.

$1.22 / $3.82 per 1M tokens

xAI's flagship for coding, knowledge work, and STEM tasks.

$2.22 / $6.65 per 1M tokens

Moonshot AI's flagship multimodal reasoning model for long-context coding and agent workflows.

$3.32 / $16.62 per 1M tokens

Z.ai's low-cost GLM 5.3 tier for high-volume chat and lightweight agents.

$0.166 / $0.554 per 1M tokens

Balanced GPT-5.6 model for everyday coding, reasoning, and agent workflows.

$2.22 / $13.29 per 1M tokens

Google's latest Flash model for fast reasoning, search, and grounding.

$0.831 / $4.15 per 1M tokens

Alibaba's flagship Qwen model for reasoning, multilingual work, and agents.

$2.22 / $6.65 per 1M tokens

DeepSeek's newest low-cost tier for high-volume chat and agent loops.

$0.155 / $0.498 per 1M tokens

Google's Flash model for fast reasoning, search, and grounding.

$0.831 / $4.15 per 1M tokens

xAI's flagship model for coding, knowledge work, and STEM tasks.

$2.22 / $6.65 per 1M tokens

Anthropic's small, fast model for high-volume work, succeeding Haiku 4.5.

$0.111 / $0.554 per 1M tokens

OpenAI's GPT-6.1 update to Sol, for demanding professional and agent work.

$2.22 / $11.08 per 1M tokens

GPT-6.1 Sol with pro-mode reasoning for higher-quality answers on complex tasks.

$2.22 / $11.08 per 1M tokens

Anthropic's Sonnet for well-scoped everyday work, succeeding Sonnet 5.

$2.22 / $11.08 per 1M tokens

Watches and listens to a video, then answers questions about it: shots, action, speech, on-screen text, lip sync.

$2.22 / $13.29 per 1M tokens

Fast, cost-efficient GPT-5.6 model for high-volume and latency-sensitive tasks.

$0.222 / $1.33 per 1M tokens

Anthropic speed/intelligence balance for production agents.

$2.22 / $11.08 per 1M tokens

Alibaba's low-cost Qwen tier for high-volume chat and lightweight agents.

$0.166 / $0.521 per 1M tokens

The GA release of DeepSeek V4 Pro, for hard coding, math, and agent workloads.

$1.25 / $3.73 per 1M tokens

Fast, cost-efficient GPT-6 model for high-volume and latency-sensitive tasks.

$0.111 / $0.554 per 1M tokens

GPT-6 Luna with pro-mode reasoning for harder tasks at Luna's price.

$0.111 / $0.554 per 1M tokens

Anthropic complex agentic coding and enterprise model.

$5.54 / $27.69 per 1M tokens

Google frontier model built for speed, search, and grounding.

$1.66 / $9.97 per 1M tokens

Google cost-efficient model for high-volume agentic tasks.

$0.277 / $1.66 per 1M tokens

OpenAI flagship model for complex reasoning and coding.

$5.54 / $33.23 per 1M tokens

DeepSeek's low-cost V4 model for high-volume chat and lightweight agents.

$0.101 / $0.202 per 1M tokens

A K2-series multimodal model for coding, UI generation, and multi-agent orchestration.

$0.693 / $3.58 per 1M tokens

OpenAI fast mini model for coding, computer use, and subagents.

$0.831 / $4.98 per 1M tokens

OpenAI lowest-cost model for high-volume lightweight tasks.

$0.222 / $1.38 per 1M tokens

OpenAI affordable frontier model for professional work.

$2.77 / $16.62 per 1M tokens

Z.ai's long-context reasoning model for complex agent and coding tasks.

$0.775 / $2.44 per 1M tokens

A multimodal K2 model for general reasoning, visual coding, and tool use.

$0.394 / $2.13 per 1M tokens

OpenAI's long-context GPT-5.2, for scripts and other long-form writing.

$1.94 / $15.51 per 1M tokens

A K2-series reasoning model for long-horizon analysis, coding, and tool calls.

$0.63 / $2.63 per 1M tokens

DeepSeek's flagship reasoning model for hard coding, math, and agent workloads.

$1.06 / $2.13 per 1M tokens

The September K2 update with a larger 262K context window and improved agentic coding.

$0.63 / $2.63 per 1M tokens

Kimi's coding-focused K2 release with persistent reasoning and long-context agent workflows.

$0.756 / $3.68 per 1M tokens

Google's newer cost-efficient tier for high-volume agentic tasks.

$0.332 / $2.77 per 1M tokens

The original K2 instruct release for coding, reasoning, and tool use.

$0.598 / $2.42 per 1M tokens

Google advanced multimodal and agentic reasoning preview.

$2.22 / $13.29 per 1M tokens

Google's established Flash model with hybrid reasoning at a low price.

$0.332 / $2.77 per 1M tokens

Perplexity's research model: searches the web, reasons over what it finds, and answers with citations.

$2.22 / $46.53 per 1M tokens

Anthropic fastest model with near-frontier intelligence.

$1.11 / $5.54 per 1M tokens

OpenAI's GPT-4o, still carried for prompts and pipelines written against it.

$2.77 / $11.08 per 1M tokens

OpenAI's faster GPT Image 2.5 tier — same 4K ceiling, less time per image.

$0.026 / image

OpenAI's newest image model — resolutions to 4K, five quality tiers, and transparent backgrounds.

$0.026 / image

OpenAI image generation and editing model.

$0.07 / image

Google high-efficiency image generation and editing.

$0.088 / image

Google ultra-low-latency image model for high-volume use.

$0.044 / image

OpenAI image model with strong text rendering.

$0.044 / image

Alibaba's higher-fidelity Qwen image tier for detail and typography.

$0.044 / image

Google professional image model for 4K visuals and text.

$0.165 / image

ByteDance flagship image generation and multi-reference editing model.

$0.05 / image

Google's latest efficient image generation and editing model, with stronger text, prompt adherence, and character consistency.

$0.088 / image

Ideogram's typography-first image model for posters, logos and designs with legible text, plus instruction editing with masks and references.

$0.066 / image

Alibaba's Qwen image model for bilingual text rendering and editing.

$0.033 / image

ByteDance's Seedream 4.7 for image generation and editing at a lower rate.

$0.033 / image

Seedream 4.7 image sets — up to 14 images that share one subject and style, from a prompt or reference images.

$0.033 / image

Midjourney V8.2, the latest release — native 2K with the high-resolution tier, through Legnext's asynchronous API.

$0.088 / image

Midjourney V8.1 image generation through Legnext's asynchronous API.

$0.095 / image

Midjourney V8 image generation through Legnext's asynchronous API.

$0.088 / image

ByteDance efficient Seedream 5 model for text-to-image and edits.

$0.039 / image

Alibaba's Wan 2.7 Pro image model, with 1K to 4K output.

$0.083 / image

Black Forest Labs' FLUX.2 Pro for photoreal generation and multi-reference editing, up to 2048px.

$0.033 / image

OpenAI lower-cost image model.

$0.022 / image

Midjourney image generation through Legnext's asynchronous API.

$0.088 / image

Google's newer Omni video model, with editing and video extension.

$0.138 / second

Google conversational video generation and editing preview.

$0.156 / second

Alibaba Wan 3.0 text-to-video with native audio, 2-30s up to 4K.

$0.20 / second

Black Forest Labs video generation with audio, 5-20s at 720p or 1080p.

$0.213 / second

ByteDance's latest Seedance model — 4-30s with native audio and reference-driven generation.

$0.35 / second

MiniMax's H3 successor to Hailuo — 4-15s at up to 2K, with a last-frame option.

$0.163 / second

ByteDance multimodal audio-video generation model.

$0.227 / second

MiniMax's top H3 tier — 5-15s, up to 4K with super-resolution.

$0.178 / second

Alibaba's highest-fidelity Wan 3.0 tier for text-to-video with audio.

$0.315 / second

Alibaba's highest-fidelity Wan 3.0 tier for image-to-video with audio.

$0.315 / second

Alibaba Wan 3.0 image-to-video with first/last frame control and audio.

$0.20 / second

ByteDance lower-cost Seedance 2.0 variant.

$0.043 / second

Alibaba Wan text-to-video with audio and multi-shot generation.

$0.125 / second

Kuaishou faster Kling 3.0 variant for production video.

$0.119 / second

Kuaishou's Kling 3.0 rendered at native 4K.

$0.446 / second

O3 rendered at native 4K.

$0.525 / second

Google latest cinematic video generation with audio.

$0.528 / second

Alibaba Wan image-to-video with keyframe and audio support.

$0.125 / second

Google efficient Veo 3.1 video generation for developer workflows.

$0.066 / second

Google faster, lower-cost Veo 3.1 variant.

$0.132 / second

ByteDance accelerated Seedance 2.0 variant for low latency.

$0.11 / second

Kuaishou latest Kling video generation model.

$0.105 / second

Kuaishou's higher-fidelity Kling 3.0 tier, rendered natively at 1080p with sound.

$0.179 / second

Kuaishou's O3 release: prompt, first frame, up to seven reference images, or an edit of an existing clip.

$0.105 / second

The higher-fidelity O3 tier, with the same modes.

$0.14 / second

Kling 3.0's motion transfer: a character image performs a reference clip's movement with steadier identity and body proportions.

$0.158 / second

Kling 3.0 motion transfer at its pro tier, the highest-fidelity Kling motion control.

$0.21 / second

Alibaba Wan fast image-to-video model.

$0.031 / second

ByteDance's Seedance 1.5 Pro — native audio, lip sync and a closing frame, at 480p or 720p.

$0.063 / second

The quicker, cheaper Seedance 1.5 Pro tier, 720p with native audio.

$0.05 / second

Alibaba Wan image-to-video with audio and multi-shot control.

$0.088 / second

Kuaishou's O1 release — 5 or 10 second clips from a prompt or a first frame.

$0.119 / second

Kling's talking avatar: one portrait and a voice track become a speaking, gesturing video as long as the audio.

$0.07 / second

Kling's talking avatar at its higher quality tier, for close-ups and longer takes.

$0.144 / second

Kling's motion transfer: the character in one image performs the dance, action or gesture of a 3–30 second reference clip.

$0.088 / second

Kling 2.6 motion transfer at its pro tier, for cleaner detail on faces and hands.

$0.14 / second

Alibaba Wan multi-shot text-to-video with audio.

$0.088 / second

MiniMax value-oriented image-to-video model.

$0.048 / second

Alibaba Wan image-to-video with audio synchronization.

$0.063 / second

ByteDance's audio-driven human video: a portrait speaks, sings and moves to the audio, with expression and body motion to match. 720p or 1080p.

$0.20 / second

Sync Labs' highest-fidelity lip-sync, keeping teeth, beards and fine facial detail intact.

$0.104 / second

Alibaba Wan text-to-video with audio synchronization.

$0.063 / second

MiniMax flagship text-to-video and image-to-video model.

$0.07 / second

Sync Labs' zero-shot lip-sync: re-times the mouth of anyone in an existing video to a new voice track.

$0.063 / second

Kling's lip-sync for short clips: re-syncs a 2–10 second video to a voice track.

$0.018 / second

MiniMax high-resolution text-to-video and image-to-video model.

$0.07 / second

MiniMax fast, natural multilingual speech, streamed as it is spoken; strong on Chinese.

$72.79 / 1M characters

MiniMax highest-fidelity multilingual speech for finished narration, streamed as it is spoken.

$121.32 / 1M characters

Suno full-song generation with vocals and instrumentation — every request returns two complete takes.

$0.152 / request

Suno's lighter, faster V6 tier for drafts and short clips — two takes per request.

$0.152 / request

Suno's experimental V6 variant with bolder, less conventional arrangements — two takes per request.

$0.152 / request

ByteDance expressive speech with sound effects, from a single script.

$0.329 / request

ElevenLabs most expressive multilingual speech model.

$103.50 / 1M characters

Google controllable multi-speaker text-to-speech.

$11.50 / 1M characters

OpenAI low-latency, steerable text-to-speech.

$13.80 / 1M characters

Google high-fidelity instrumental music generation.

$0.069 / second

ElevenLabs low-latency speech for realtime agents.

$51.75 / 1M characters

OpenAI fast, low-cost speech-to-text for dictation and voice input.

$0.0000607 / second

OpenAI most accurate speech-to-text, for noisy audio, accents and mixed languages.

$0.000121 / second

Alibaba Qwen speech-to-text, strong on Chinese, its dialects and English, for dictation.

$0.0000425 / second

ByteDance controllable music generation and editing.

$0.034 / second

Voyage AI general-purpose multilingual embeddings for retrieval, 1024 dimensions by default.

$0.066 per 1M input tokens

Voyage AI low-cost multilingual embeddings for retrieval, in the same vector space as Voyage 4.

$0.022 per 1M input tokens

Alibaba Qwen multilingual text embeddings, 100+ languages and strong on Chinese, with a vector length you choose.

$0.011 per 1M input tokens

BAAI open multilingual text embeddings, 1024 dimensions, for retrieval across 100+ languages.

$0.011 per 1M input tokens

OpenAI small, low-cost text embeddings for search and retrieval, with a vector length you choose.

$0.022 per 1M input tokens

OpenAI most capable text embeddings, multilingual, with a vector length you choose.

$0.144 per 1M input tokens

ByteDance's Seedream 5.0 Pro layer decomposition — a base image plus up to 16 transparent layers, each with its name, z-order, and bounding box.

$0.037 / layer

Alibaba's open layer-decomposition model — splits one image into 2 to 10 editable RGBA layers.

$0.028 / layer

Re-shoots one image from another camera angle — orbit, elevation and zoom — keeping the subject and scene.

$0.04 / image

ByteDance's temporally consistent depth estimation — turns any video into a flicker-free grayscale depth video.

$0.063 / second