52 live models • One key • Flat ₹99 per 1M tokens
Every model, best first.
The full live catalog behind your single key — flagship reasoning, agentic coders, vision, ultra-fast flash, embeddings and more providers. Pick a lane below; the router leans toward the healthiest model in it with automatic failover.
64 models
deepseek-v4-pro
DeepSeek AI • frontier reasoning
deepseek-v4-pro-0813
DeepSeek AI • frontier reasoning
deepseek-v4-flash
DeepSeek AI • ultra-low latency
deepseek-v3.2
DeepSeek AI • efficient reasoning
mercury-2.5
Inception Labs • diffusion LLM • ultra-fast
mercury-2
Inception Labs • diffusion LLM • high speed
ernie-5.1
Baidu ERNIE • knowledge reasoning
ernie-5.0
Baidu ERNIE • flagship reasoning
qwen3.8-max
Alibaba Qwen • frontier reasoning
qwen3.7-max
Alibaba Qwen • frontier reasoning
qwen-max
Alibaba Qwen • frontier reasoning
qwen3.8-2.4t-a95b
Alibaba Qwen • frontier reasoning
qwq-plus
Alibaba Qwen • deep reasoning
kimi-k3
Moonshot Kimi • long-context reasoning
glm-5.2
Zhipu GLM • frontier reasoning
qwen3.6-max-preview
Alibaba Qwen • frontier reasoning
qwen3.7-plus
Alibaba Qwen • balanced flagship
qwen3.6-plus
Alibaba Qwen • balanced flagship
qwen3.5-plus
Alibaba Qwen • balanced flagship
qwen-plus
Alibaba Qwen • balanced flagship
qwen3.8-flash
Alibaba Qwen • ultra-low latency
qwen3.7-flash
Alibaba Qwen • ultra-low latency
qwen3.6-flash
Alibaba Qwen • ultra-low latency
qwen3.5-flash
Alibaba Qwen • ultra-low latency
qwen-flash
Alibaba Qwen • ultra-low latency
qwen-turbo
Alibaba Qwen • high-speed turbo
qwen3.6-27b
Alibaba Qwen • efficient 27B
qwen3.5-27b
Alibaba Qwen • efficient 27B
qwen3-coder-plus
Alibaba Qwen • agentic coding
kimi-k2.7-code
Moonshot Kimi • agentic coding
qwen-coder-plus
Alibaba Qwen • agentic coding
qwen3-coder-next
Alibaba Qwen • agentic coding
qwen3-coder-480b-a35b-instruct
Alibaba Qwen • MoE coding giant
qwen3-coder-30b-a3b-instruct
Alibaba Qwen • lightweight coder
qwen3-coder-flash
Alibaba Qwen • fast coding
qwen3-vl-plus
Alibaba Qwen • multimodal vision
qwen-vl-max
Alibaba Qwen • flagship vision
qwen-vl-plus
Alibaba Qwen • multimodal vision
qwen3-vl-flash
Alibaba Qwen • fast vision
qwen-vl-ocr
Alibaba Qwen • document OCR
qwen3.5-omni-plus
Alibaba Qwen • omni multimodal
qwen3.5-omni-flash
Alibaba Qwen • fast omni
gemini-2.5-pro
Google • flagship multimodal
gemini-2.5-flash
Google • fast multimodal
gemini-2.0-flash
Google • fast multimodal
gemini-2.0-flash-lite
Google • lightest flash
llama-3.3-70b-versatile
Meta • via Groq, ultra-fast
llama-3.1-8b-instant
Meta • via Groq, instant
mixtral-8x7b-32768
Mistral AI • via Groq, MoE
llama3.1-70b
Meta • via Cerebras, wafer-speed
llama3.1-8b
Meta • via Cerebras, wafer-speed
stable-diffusion-xl
Stability AI / Baidu • photorealistic diffusion
text-embedding-v4
Alibaba Qwen • 1536/1024-dim dense vector
text-embedding-v3
Alibaba Qwen • 1024-dim dense vector
qwen3.7-text-embedding
Alibaba Qwen • vector embeddings
bge-large-zh
BAAI / Baidu • high-precision embeddings
bge-large-en
BAAI / Baidu • English embeddings
tao-8k
Baidu Qianfan • 8K-context embeddings
embedding-v1
Baidu Qianfan • semantic embeddings
m3e-base
Baidu Qianfan • multi-lingual embeddings
wanx2.1-t2v-turbo
Alibaba Wan • text-to-video turbo
wanx2.1-t2v-plus
Alibaba Wan • cinema HD video
sora-1.0
OpenAI Sora • motion video
cogvideox-5b
Zhipu AI • high-res video generator
Need a model that's not listed? We wire AI models on demand.
From the latest Google Gemini drops to ultra-low-latency diffusion & image models to Nvidia NIM builds — tell us the model, we plug it into your same key at the same flat ₹99/1M rate.
Gemini latest
Diffusion • Flux • SDXL-Turbo
Nvidia NIM
Whisper • TTS • custom