Skip to content
Rapid Router
All models
Hf

Hf API pricing in India

82 Hf models live through one API key and one balance, priced per million tokens in rupees.

Models live
82
Lowest input
₹0.987/M
Max context
—

Every Hf model below is callable right now through Rapid Router with one API key. The price next to each model is the provider's list price per million tokens, shown in rupees at today's billing rate (the market rate plus our currency margin); in dollars it is exactly what your balance is debited. Our platform fee is charged once, when you buy credits, so there is nothing added per request.

To use a Hf model, point the OpenAI SDK (or any OpenAI-compatible client) at https://rapidrouter.ai/api/v1, set the model string to the id shown in grey, and send the same request you send today. Top up with UPI or a card and a GST invoice follows every recharge.

To weigh a Hf model against Claude, GPT, Gemini or any other family, the model rankings put independent LMArena ratings next to each price, and the cost calculator works out what your own monthly token volume would cost on every model.

DeepSeek R1:novita

₹69.12 in · ₹247 out /1M

hf/deepseek-ai/deepseek-r1:novita, by Hf
DeepSeek V3.1 Terminus:deepinfra

₹26.66 in · ₹93.80 out /1M

hf/deepseek-ai/deepseek-v3.1-terminus:deepinfra, by Hf
DeepSeek V3.2:deepinfra

₹25.67 in · ₹37.52 out /1M

hf/deepseek-ai/deepseek-v3.2:deepinfra, by Hf
DeepSeek V3:deepinfra

₹31.60 in · ₹87.88 out /1M

hf/deepseek-ai/deepseek-v3:deepinfra, by Hf
DeepSeek V4 Flash 0731:deepinfra

₹5.92 in · ₹17.77 out /1M

hf/deepseek-ai/deepseek-v4-flash-0731:deepinfra, by Hf
DeepSeek V4 Flash Vision Exp:deepinfra

₹43.45 in · ₹130 out /1M

hf/deepseek-ai/deepseek-v4-flash-vision-exp:deepinfra, by Hf
DeepSeek V4 Flash:deepinfra

₹8.89 in · ₹17.77 out /1M

hf/deepseek-ai/deepseek-v4-flash:deepinfra, by Hf
DeepSeek V4 Pro:deepinfra

₹128 in · ₹257 out /1M

hf/deepseek-ai/deepseek-v4-pro:deepinfra, by Hf
DeepSeek V4.1 Flash 1017:novita

₹29.62 in · ₹118 out /1M

hf/deepseek-ai/deepseek-v4.1-flash-1017:novita, by Hf
DeepSeek V4.1 Flash:deepinfra

₹19.75 in · ₹59.24 out /1M

hf/deepseek-ai/deepseek-v4.1-flash:deepinfra, by Hf
Gemma 3 12B It:deepinfra

₹4.94 in · ₹14.81 out /1M

hf/google/gemma-3-12b-it:deepinfra, by Hf
Gemma 3 27B It:deepinfra

₹7.90 in · ₹15.80 out /1M

hf/google/gemma-3-27b-it:deepinfra, by Hf
Gemma 3 4B It:deepinfra

₹4.94 in · ₹9.87 out /1M

hf/google/gemma-3-4b-it:deepinfra, by Hf
Gemma 4 26B A4B It:deepinfra

₹6.91 in · ₹33.57 out /1M

hf/google/gemma-4-26b-a4b-it:deepinfra, by Hf
Gemma 4 31B It:novita

₹13.82 in · ₹39.50 out /1M

hf/google/gemma-4-31b-it:novita, by Hf
Granite 4.2 30b:deepinfra

₹15.80 in · ₹64.18 out /1M

hf/ibm-granite/granite-4.2-30b:deepinfra, by Hf
Granite 4.2 3b:deepinfra

₹2.96 in · ₹11.85 out /1M

hf/ibm-granite/granite-4.2-3b:deepinfra, by Hf
Granite 4.2 8b:deepinfra

₹5.92 in · ₹24.68 out /1M

hf/ibm-granite/granite-4.2-8b:deepinfra, by Hf
Ling 3.0 Flash Fin:deepinfra

₹5.92 in · ₹17.77 out /1M

hf/inclusionai/ling-3.0-flash-fin:deepinfra, by Hf
Ling 3.0 Flash:novita

₹5.92 in · ₹17.77 out /1M

hf/inclusionai/ling-3.0-flash:novita, by Hf
Ling 3.1 Flash:novita

₹29.62 in · ₹88.87 out /1M

hf/inclusionai/ling-3.1-flash:novita, by Hf
Llama 3.1 8B Instruct:novita

₹1.97 in · ₹4.94 out /1M

hf/meta-llama/llama-3.1-8b-instruct:novita, by Hf
Llama 3.3 70B Instruct:novita

₹13.33 in · ₹39.50 out /1M

hf/meta-llama/llama-3.3-70b-instruct:novita, by Hf
Llama 4 Scout 17B 16e Instruct:nscale

₹8.89 in · ₹28.63 out /1M

hf/meta-llama/llama-4-scout-17b-16e-instruct:nscale, by Hf
Llama Guard 4 12b:deepinfra

₹17.77 in · ₹17.77 out /1M

hf/meta-llama/llama-guard-4-12b:deepinfra, by Hf
Muse Glimmer 30b:deepinfra

₹29.62 in · ₹118 out /1M

hf/meta-models/muse-glimmer-30b:deepinfra, by Hf
Muse Spark Open 816B A42b:baseten

₹123 in · ₹420 out /1M

hf/meta-models/muse-spark-open-816b-a42b:baseten, by Hf
Phi 4:deepinfra

₹6.91 in · ₹13.82 out /1M

hf/microsoft/phi-4:deepinfra, by Hf
MiniMax M2.5:novita

₹29.62 in · ₹118 out /1M

hf/minimaxai/minimax-m2.5:novita, by Hf
MiniMax M2.7:novita

₹29.62 in · ₹118 out /1M

hf/minimaxai/minimax-m2.7:novita, by Hf
MiniMax M3:novita

₹29.62 in · ₹118 out /1M

hf/minimaxai/minimax-m3:novita, by Hf
Kimi K2.5:deepinfra

₹44.43 in · ₹222 out /1M

hf/moonshotai/kimi-k2.5:deepinfra, by Hf
Kimi K2.6:deepinfra

₹74.05 in · ₹346 out /1M

hf/moonshotai/kimi-k2.6:deepinfra, by Hf
Kimi K2.7 Code:deepinfra

₹67.14 in · ₹336 out /1M

hf/moonshotai/kimi-k2.7-code:deepinfra, by Hf
Kimi K3:deepinfra

₹281 in · ₹1,407 out /1M

hf/moonshotai/kimi-k3:deepinfra, by Hf
Nvidia Nemotron 3 Ultra 550B A55B Bf16:deepinfra

₹98.74 in · ₹494 out /1M

hf/nvidia/nvidia-nemotron-3-ultra-550b-a55b-bf16:deepinfra, by Hf
Nvidia Nemotron 3.5 Super VL 120B A12B Nvfp4:baseten

₹19.75 in · ₹49.37 out /1M

hf/nvidia/nvidia-nemotron-3.5-super-vl-120b-a12b-nvfp4:baseten, by Hf
GPT OSS 120b:deepinfra

₹3.65 in · ₹16.79 out /1M

hf/openai/gpt-oss-120b:deepinfra, by Hf
GPT OSS 20b:deepinfra

₹2.96 in · ₹13.82 out /1M

hf/openai/gpt-oss-20b:deepinfra, by Hf
Qwen2.5 72B Instruct:deepinfra

₹35.55 in · ₹39.50 out /1M

hf/qwen/qwen2.5-72b-instruct:deepinfra, by Hf
Qwen2.5 Coder 32B Instruct:nscale

₹5.92 in · ₹19.75 out /1M

hf/qwen/qwen2.5-coder-32b-instruct:nscale, by Hf
Qwen2.5 Coder 7B Instruct:nscaleLowest cost

₹0.987 in · ₹2.96 out /1M

hf/qwen/qwen2.5-coder-7b-instruct:nscale, by Hf
Qwen2.5 VL 72B Instruct:ovhcloud

₹99.73 in · ₹99.73 out /1M

hf/qwen/qwen2.5-vl-72b-instruct:ovhcloud, by Hf
Qwen3 14b:nscale

₹6.91 in · ₹19.75 out /1M

hf/qwen/qwen3-14b:nscale, by Hf
Qwen3 235B A22B Thinking 2507:deepinfra

₹22.71 in · ₹227 out /1M

hf/qwen/qwen3-235b-a22b-thinking-2507:deepinfra, by Hf
Qwen3 30B A3b:deepinfra

₹11.85 in · ₹49.37 out /1M

hf/qwen/qwen3-30b-a3b:deepinfra, by Hf
Qwen3 32b:nscale

₹7.90 in · ₹24.68 out /1M

hf/qwen/qwen3-32b:nscale, by Hf
Qwen3 4B Instruct 2507:nscaleLowest cost

₹0.987 in · ₹2.96 out /1M

hf/qwen/qwen3-4b-instruct-2507:nscale, by Hf
Qwen3 8b:nscale

₹6.91 in · ₹17.77 out /1M

hf/qwen/qwen3-8b:nscale, by Hf
Qwen3 Next 80B A3B Instruct:deepinfra

₹8.89 in · ₹109 out /1M

hf/qwen/qwen3-next-80b-a3b-instruct:deepinfra, by Hf
Qwen3 VL 235B A22B Instruct:deepinfra

₹19.75 in · ₹86.89 out /1M

hf/qwen/qwen3-vl-235b-a22b-instruct:deepinfra, by Hf
Qwen3 VL 30B A3B Instruct:deepinfra

₹14.81 in · ₹59.24 out /1M

hf/qwen/qwen3-vl-30b-a3b-instruct:deepinfra, by Hf
Qwen3.5 122B A10b:deepinfra

₹28.63 in · ₹237 out /1M

hf/qwen/qwen3.5-122b-a10b:deepinfra, by Hf
Qwen3.5 27b:deepinfra

₹25.67 in · ₹257 out /1M

hf/qwen/qwen3.5-27b:deepinfra, by Hf
Qwen3.5 35B A3b:deepinfra

₹13.82 in · ₹98.74 out /1M

hf/qwen/qwen3.5-35b-a3b:deepinfra, by Hf
Qwen3.5 397B A17b:deepinfra

₹44.43 in · ₹296 out /1M

hf/qwen/qwen3.5-397b-a17b:deepinfra, by Hf
Qwen3.5 9b:deepinfra

₹9.87 in · ₹14.81 out /1M

hf/qwen/qwen3.5-9b:deepinfra, by Hf
Qwen3.6 35B A3b:deepinfra

₹9.87 in · ₹93.80 out /1M

hf/qwen/qwen3.6-35b-a3b:deepinfra, by Hf
Qwen3.8 2.4t A95b:deepinfra

₹197 in · ₹592 out /1M

hf/qwen/qwen3.8-2.4t-a95b:deepinfra, by Hf
Qwen3.8 27b:deepinfra

₹19.75 in · ₹247 out /1M

hf/qwen/qwen3.8-27b:deepinfra, by Hf
Beam 1.0 501B A23b:baseten

₹39.50 in · ₹118 out /1M

hf/reflection/beam-1.0-501b-a23b:baseten, by Hf
L3 8B Stheno V3.2:novita

₹4.94 in · ₹4.94 out /1M

hf/sao10k/l3-8b-stheno-v3.2:novita, by Hf
Bielik 11B V3.0 Instruct:publicai

₹39.50 in · ₹39.50 out /1M

hf/speakleash/bielik-11b-v3.0-instruct:publicai, by Hf
Apertus 8B Instruct 2509:publicai

₹9.87 in · ₹19.75 out /1M

hf/swiss-ai/apertus-8b-instruct-2509:publicai, by Hf
Apertus V1.5 70b:publicai

₹80.97 in · ₹288 out /1M

hf/swiss-ai/apertus-v1.5-70b:publicai, by Hf
Hy3:deepinfra

₹12.84 in · ₹52.33 out /1M

hf/tencent/hy3:deepinfra, by Hf
Hy4 Preview:novita

₹82.35 in · ₹247 out /1M

hf/tencent/hy4-preview:novita, by Hf
Inkling Small:deepinfra

₹44.43 in · ₹118 out /1M

hf/thinkingmachines/inkling-small:deepinfra, by Hf
Inkling:deepinfra

₹93.80 in · ₹400 out /1M

hf/thinkingmachines/inkling:deepinfra, by Hf
MiMo V2.5 Pro:deepinfra

₹98.74 in · ₹296 out /1M

hf/xiaomimimo/mimo-v2.5-pro:deepinfra, by Hf
MiMo V2.6 Flash Rl:novita

₹13.82 in · ₹27.65 out /1M

hf/xiaomimimo/mimo-v2.6-flash-rl:novita, by Hf
MiMo V2.6 Pro Rl:novita

₹42.95 in · ₹85.90 out /1M

hf/xiaomimimo/mimo-v2.6-pro-rl:novita, by Hf
GLM 4.5 Air:novita

₹12.84 in · ₹83.93 out /1M

hf/zai-org/glm-4.5-air:novita, by Hf
GLM 4.5v:novita

₹59.24 in · ₹178 out /1M

hf/zai-org/glm-4.5v:novita, by Hf
GLM 4.6:deepinfra

₹49.37 in · ₹197 out /1M

hf/zai-org/glm-4.6:deepinfra, by Hf
GLM 4.7 Flash:deepinfra

₹5.92 in · ₹39.50 out /1M

hf/zai-org/glm-4.7-flash:deepinfra, by Hf
GLM 4.7:deepinfra

₹39.50 in · ₹173 out /1M

hf/zai-org/glm-4.7:deepinfra, by Hf
GLM 5.1:deepinfra

₹104 in · ₹346 out /1M

hf/zai-org/glm-5.1:deepinfra, by Hf
GLM 5.2:deepinfra

₹74.05 in · ₹237 out /1M

hf/zai-org/glm-5.2:deepinfra, by Hf
GLM 5.3 Flash:novita

₹14.81 in · ₹49.37 out /1M

hf/zai-org/glm-5.3-flash:novita, by Hf
GLM 5.3:deepinfra

₹88.87 in · ₹395 out /1M

hf/zai-org/glm-5.3:deepinfra, by Hf
GLM 5:deepinfra

₹59.24 in · ₹205 out /1M

hf/zai-org/glm-5:deepinfra, by Hf

Hf API: common questions

How do I get a Hf API key in India?
Create a Rapid Router account, open API keys in the dashboard and create a key. That one key calls every Hf model, plus Claude, GPT, Gemini and 600+ others, through the OpenAI SDK with the base URL https://rapidrouter.ai/api/v1. You do not need a separate Hf account or an international card.
Can I pay for the Hf API with UPI?
Yes. Top up your Rapid Router balance in rupees with UPI or an Indian card, and every recharge comes with a GST invoice. Hf requests are then billed from that balance at Hf's list price per million tokens.
What does the Hf API cost?
Input prices run from ₹0.987 per million tokens (Qwen2.5 Coder 7B Instruct:nscale) to ₹281 (Kimi K3:deepinfra). Each model's input and output price is in the table on this page.
Is there a free Hf API?
Not right now: every Hf model here is billed at its list price. The daily free allowance on Rapid Router covers the models marked Free in the catalogue.
Which Hf model should I use?
Start with the cheapest model that handles your task and move up only for the requests it gets wrong. The best model rankings compare Hf models with every other family on independent votes and price.