Free tool
AI API cost calculator
Enter how many tokens your product uses in a month and see what every model costs, side by side. Prices are the live catalog prices Rapid Router bills at, in US dollars.
| Model | Input / 1M | Output / 1M | Per month (USD) |
|---|---|---|---|
| Llama 3.2 1B InstructMeta | $0.01 | $0.01 | $0.07 |
| Qwen3.5 0.8BQwen | $0.01 | $0.05 | $0.20 |
| Llama 3.2 3B InstructMeta | $0.02 | $0.02 | $0.24 |
| Meta Llama 3.1 8B Instruct TurboMeta | $0.02 | $0.04 | $0.28 |
| Meta Llama 3.1 8B Instruct Turbo (free)Meta | $0.02 | $0.04 | $0.28 |
| Llama-3.1-8B-InstructMeta | $0.02 | $0.05 | $0.30 |
| Llama-3.1-8B-InstructNovita | $0.02 | $0.05 | $0.30 |
| Qwen2.5 Coder 7BQwen | $0.03 | $0.05 | $0.35 |
| Mistral NemoMistral | $0.03 | $0.03 | $0.35 |
| Mistral Nemo (free)Mistral | $0.03 | $0.03 | $0.35 |
| DeepSeek OCR 2Novita | $0.03 | $0.03 | $0.36 |
| Meta Llama 3 8B InstructMeta | $0.03 | $0.04 | $0.38 |
| mistral-7b-instruct-v0.3Mistral | $0.03 | $0.05 | $0.39 |
| Gemma 4 E4B ITGoogle | $0.02 | $0.10 | $0.40 |
| qwen3.5-2bQwen | $0.02 | $0.10 | $0.40 |
| Gemma 2 9BGoogle | $0.03 | $0.06 | $0.42 |
| L3 8B Lunaris V1Sao10k | $0.03 | $0.06 | $0.42 |
| DeepSeek OCRDeepseek Ai | $0.03 | $0.10 | $0.50 |
| L3 8B Lunaris V1 TurboSao10k | $0.04 | $0.05 | $0.50 |
| Qwen3.7 FlashQwen | $0.03 | $0.13 | $0.56 |
| Qwen3.7 FlashQwen | $0.03 | $0.13 | $0.56 |
| GPT OSS 20BOpenAI | $0.03 | $0.14 | $0.58 |
| GPT OSS 20B (free)OpenAI | $0.03 | $0.14 | $0.58 |
| Sao10K Stheno 8bNovita | $0.05 | $0.05 | $0.60 |
| Qwen3.5 4BQwen | $0.03 | $0.15 | $0.60 |
| Qwen2.5 7B InstructQwen | $0.04 | $0.10 | $0.60 |
| AutoGLM Phone 9B MultilingualNovita | $0.04 | $0.14 | $0.63 |
| Llama-Guard-3-8B (Deep Infra)Meta | $0.06 | $0.06 | $0.66 |
| WizardLM 2 7BMicrosoft | $0.06 | $0.06 | $0.66 |
| Mistral 7B Instruct V0.1Mistral | $0.06 | $0.06 | $0.66 |
| mistral-7b-instruct-v0.2Mistral | $0.06 | $0.06 | $0.66 |
| Mistral Small 3Mistral | $0.05 | $0.08 | $0.66 |
| Mistral Small 3 (free)Mistral | $0.05 | $0.08 | $0.66 |
| OpenChat 3.6 8BOpenchat | $0.06 | $0.06 | $0.66 |
| OpenChat 3.5Openchat | $0.06 | $0.06 | $0.66 |
| Qwen2 7B InstructQwen | $0.06 | $0.06 | $0.66 |
| Gemma 3 4B ITGoogle | $0.05 | $0.10 | $0.70 |
| Gemma 3 4B IT (free)Google | $0.05 | $0.10 | $0.70 |
| Phi 4 MultimodalMicrosoft | $0.05 | $0.10 | $0.70 |
| Mistral Small 3.1 24BMistral | $0.05 | $0.10 | $0.70 |
| GPT OSS 120BOpenAI | $0.04 | $0.17 | $0.71 |
| nvidia-nemotron-nano-9b-v2NVIDIA | $0.04 | $0.16 | $0.72 |
| Gemma 3 12B ITGoogle | $0.05 | $0.15 | $0.80 |
| Devstral Small 2505Mistral | $0.06 | $0.12 | $0.84 |
| CodeGemma 7BGoogle | $0.07 | $0.07 | $0.84 |
| Gemma 1.1 7BGoogle | $0.07 | $0.07 | $0.84 |
| Qwen2.5-Coder-32B-InstructQwen | $0.06 | $0.15 | $0.90 |
| Nemotron 3 Nano 30B A3BNVIDIA | $0.05 | $0.20 | $0.90 |
| Nvidia Nemotron 3.5 LightningNVIDIA | $0.06 | $0.16 | $0.92 |
| Ling 3.0 Flash VLDeepinfra | $0.06 | $0.18 | $0.96 |
| DeepSeek V4 Flash 0731DeepSeek | $0.06 | $0.18 | $0.96 |
| Ling 3.0 FlashInclusionai | $0.06 | $0.18 | $0.96 |
| Phi 4Microsoft | $0.07 | $0.14 | $0.98 |
| Qwen-VL OCRQwen | $0.07 | $0.16 | $1.02 |
| Granite 4.2 8BDeepinfra | $0.06 | $0.25 | $1.10 |
| Gemma 3 27B ITGoogle | $0.08 | $0.16 | $1.12 |
| Mistral-Small-3.2-24B-Instruct-2506Mistral | $0.07 | $0.20 | $1.15 |
| Qwen3.5 FlashQwen | $0.07 | $0.26 | $1.17 |
| Qwen3.5 FlashQwen | $0.07 | $0.26 | $1.17 |
| Ling 3.0 Flash FinNovita | $0.07 | $0.22 | $1.19 |
| Ling 3.0 Flash SanteNovita | $0.07 | $0.22 | $1.19 |
| Ling 3.0 Flash VLNovita | $0.07 | $0.22 | $1.19 |
| Qwen 2 1.5B InstructArize Ai | $0.10 | $0.10 | $1.20 |
| Llama2 13B TiefighterKoboldai | $0.10 | $0.10 | $1.20 |
| Ministral 3 3BMistral | $0.10 | $0.10 | $1.20 |
| Ministral 3B (latest)Mistral | $0.10 | $0.10 | $1.20 |
| Qwen3-Coder 30B-A3B InstructQwen | $0.07 | $0.26 | $1.22 |
| DeepSeek V4 FlashDeepSeek | $0.09 | $0.18 | $1.26 |
| Devstral SmallMistral | $0.07 | $0.28 | $1.26 |
| Olmocr 2 7B 1025Allenai | $0.09 | $0.19 | $1.28 |
| GPT 5 NanoOpenAI | $0.05 | $0.40 | $1.30 |
| GPT 5 NanoOpenAI | $0.05 | $0.40 | $1.30 |
| Qwen3.5 9BQwen | $0.10 | $0.15 | $1.30 |
| GPT OSS Safeguard 20BOpenAI | $0.07 | $0.30 | $1.35 |
| Qwen3 32BQwen | $0.08 | $0.28 | $1.36 |
| Qwen3 32B (free)Qwen | $0.08 | $0.28 | $1.36 |
| Gemma 4 26B A4B ITGoogle | $0.07 | $0.34 | $1.38 |
| Phi-4-reasoning-plusMicrosoft | $0.07 | $0.35 | $1.40 |
| GLM-4.7-FlashZ.ai | $0.06 | $0.40 | $1.40 |
| Step 3.5 FlashStepfun | $0.09 | $0.30 | $1.50 |
| Chronos Hermes 13B V2Austism | $0.13 | $0.13 | $1.56 |
| MythoMax L2 13B TurboGryphe | $0.13 | $0.13 | $1.56 |
| Llama 2 13B Chat HfMeta | $0.13 | $0.13 | $1.56 |
| Sky T1 32B PreviewNovasky Ai | $0.12 | $0.18 | $1.56 |
| QwQ 32BQwen | $0.12 | $0.18 | $1.56 |
| Gemma 4 31B It TurboGoogle | $0.09 | $0.34 | $1.58 |
| Llama 4 Scout 17BMeta | $0.10 | $0.30 | $1.60 |
| Llama 4 Scout 17B (free)Meta | $0.10 | $0.30 | $1.60 |
| Llama 3.3 70B TurboDeepinfra | $0.10 | $0.32 | $1.64 |
| nvidia-nemotron-3-super-120b-a12bNVIDIA | $0.09 | $0.40 | $1.65 |
| nvidia-nemotron-3-super-120b-a12b (free)NVIDIA | $0.09 | $0.40 | $1.65 |
| Qwen3 14BQwen | $0.12 | $0.24 | $1.68 |
| Qwen3 14B (free)Qwen | $0.12 | $0.24 | $1.68 |
| Phi 3 Medium 4K InstructMicrosoft | $0.14 | $0.14 | $1.68 |
| Starcoder2 15B Instruct V0.1Bigcode | $0.15 | $0.15 | $1.80 |
| Seed 2.0 MiniDeepinfra | $0.10 | $0.40 | $1.80 |
| Ministral 3 8BMistral | $0.15 | $0.15 | $1.80 |
| Ministral 8B (latest)Mistral | $0.15 | $0.15 | $1.80 |
| Voxtral SmallMistral | $0.10 | $0.40 | $1.80 |
| Voxtral Small (latest)Mistral | $0.10 | $0.40 | $1.80 |
| GPT 4.1 NanoOpenAI | $0.10 | $0.40 | $1.80 |
| GPT 4.1 NanoOpenAI | $0.10 | $0.40 | $1.80 |
| Qwen3.8 FlashDeepinfra | $0.11 | $0.38 | $1.89 |
| Qwen Mt LiteQwen | $0.12 | $0.36 | $1.92 |
| GLM-5.3-FlashZ.ai | $0.01 | $0.90 | $1.92 |
| MiMo-V2.6-FlashDeepinfra | $0.14 | $0.28 | $1.96 |
| MiMo-V2.6-FlashNovita | $0.14 | $0.28 | $1.96 |
| Claude Haiku 5.5Anthropic | $0.10 | $0.50 | $2.00 |
| GPT 6 LunaOpenAI | $0.10 | $0.50 | $2.00 |
| Qwen3 235B-A22B Instruct 2507Qwen | $0.09 | $0.55 | $2.00 |
| Llama Guard 4 12BMeta | $0.18 | $0.18 | $2.16 |
| Qwen3 30B A3BQwen | $0.12 | $0.50 | $2.20 |
| Qwen3 30B A3B (free)Qwen | $0.12 | $0.50 | $2.20 |
| Qwen3 VL 4B InstructQwen | $0.10 | $0.60 | $2.20 |
| QwQ 32BQwen | $0.15 | $0.40 | $2.30 |
| Qwen3 30B A3B Instruct 2507Qwen | $0.13 | $0.52 | $2.34 |
| Ministral 3 14BMistral | $0.20 | $0.20 | $2.40 |
| Ministral 14BMistral | $0.20 | $0.20 | $2.40 |
| Nemotron Content Safety 3.5NVIDIA | $0.20 | $0.20 | $2.40 |
| Qwen3.8 FlashNovita | $0.15 | $0.47 | $2.44 |
419 paid chat models (free-tier models are left out; see the free tier in the docs). Platform fee of is charged once when you buy credits, not per request. Image, video and speech models are priced per unit on their own pages.
Common questions
How is the monthly cost calculated?
Input tokens per month × the model's input price per million, plus output tokens × the output price per million. Prices are the provider list prices Rapid Router bills at, in US dollars. The platform fee is charged once when you buy credits and is shown separately.
How many tokens is a page of text?
Roughly 1,300 tokens per 1,000 English words. Languages in non-Latin scripts tokenise less efficiently on most models, often 2 to 3 times more tokens for the same text.
Why do output tokens cost more than input?
Every provider prices generation higher than reading, usually 3 to 5 times. A workload that reads a lot and writes a little (search, classification, RAG) is cheaper per token than one that writes long answers.
Does the fee change the per-request price?
No. The per-request price is exactly the catalog price. The platform fee is added once when you buy credits, so the number this calculator shows is what your balance is debited.
Ready to try one? Create a key, point your SDK at https://rapidrouter.ai/api/v1, and the first requests are on the free daily allowance.