Live · synced nightly

Every model.
One API key.

Prices are honest list prices — per million tokens, in USD. Context windows shown in tokens. Capabilities are inherited from the upstream provider; we never gate features behind tiers.

303
models
19
providers
$0.03
cheapest · per M in
1.1M
largest context
Type
303 models
GPT family + reasoning

OpenAI82 models

openai.com
Gemini — multimodal, 1M context

Google34 models

deepmind.google/technologies/gemini
Qwen — multilingual frontier

Alibaba45 models

qwenlm.github.io
M-series — fast and cheap

MiniMax10 models

minimax.com
GLM — open weights frontier

Z.AI17 models

zhipuai.cn
Claude — long context, agentic

Anthropic18 models

anthropic.com
Seedance + Seedream — generative media

ByteDance14 models

bytedance.com
MiMo — long-context Chinese frontier

Xiaomi5 models

www.mimo.ai
Cost-leader reasoning

DeepSeek11 models

deepseek.com
Cinematic text-to-video

KlingAI8 models

klingai.com
Vector + raster image generation

Recraft9 models

recraft.ai
Compact reasoning

Arcee AI1 model

arcee.ai
Mercury — diffusion-based LLM

Inception3 models

www.inceptionlabs.ai
Kimi — long-context generalist

Moonshot AI8 models

www.moonshot.cn
Nemotron — open weights

NVIDIA6 models

www.nvidia.com
cohere

cohere5 models

meta

meta12 models

mistral

mistral10 models

perplexity

perplexity5 models

Ready

Pick a model and ship a request in three minutes.

Every model on this page is reachable from the same https://synapse.garden/api/v1 base URL. No per-provider auth. No region juggling.