1 · HOOK
10 AI APIs ranked by ₹ per million tokens
Pricing data from OpenRouter, refreshed weekly.
2 · CONTEXT
What you'll find
Pricing data from OpenRouter, refreshed weekly. Sorted to surface the most useful first.
3 · NUMBER
FREE
Mistral: Mistral Nemo — per million tokens
4 · QUOTE
"Meta's latest class of model (Llama 3"
— Meta: Llama 3.1 8B Instruct, meta-llama
5 · MYTH
Frontier AI is always expensive.
Actually: Mistral: Mistral Nemo: A 12B parameter model with a 128k token context length built by Mistral in collaboration with NVIDIA. The model is multi
6 · NUMBER2
8K
Meta: Llama 3 8B Instruct — context window
7 · QUOTE
"Granite-4"
— IBM: Granite 4.0 Micro, ibm-granite
8 · COMPARE
Sao10K: Llama 3 8B Lunaris vs GPT-5.5 Pro
Sao10K: Llama 3 8B Lunaris: ₹0/M
GPT-5.5 Pro: ₹2,505/M
9 · TRUE BUT INCOMPLETE
Price isn't everything — latency, India availability, and Indic support shift the math.
10 · CLOSER
Full pricing table
Full table: /explore/api-comparison/