AI Model API Pricing: What a Request Really Costs
141 models, 16 labs, prices synced every 6 hours. Set your prompt size below — we'll rank every model by what that request actually costs. No affiliate links on this page.
Token cost calculator
Rule of thumb: 1 token ≈ 4 characters ≈ ¾ of an English word. List prices; caching and batch discounts can lower real costs.
| # | Model | Lab | This request | Input $/1M | Output $/1M | Context |
|---|---|---|---|---|---|---|
| 1 | Llama 3.1 8B Instruct | Meta | $0.0011 | $0.02 | $0.03 | 131k |
| 2 | Llama 3.2 1B Instruct | Meta | $0.0018 | $0.027 | $0.201 | 131k |
| 3 | Nova Micro 1.0 | Amazon | $0.0020 | $0.035 | $0.14 | 128k |
| 4 | Command R7B (12-2024) | Cohere | $0.0022 | $0.0375 | $0.15 | 128k |
| 5 | Nemotron 3 Nano 30B A3B | NVIDIA | $0.0029 | $0.05 | $0.2 | 262k |
| 6 | Llama 3.2 3B Instruct | Meta | $0.0032 | $0.05 | $0.33 | 131k |
| 7 | Nova Lite 1.0 | Amazon | $0.0035 | $0.06 | $0.24 | 300k |
| 8 | Gemma 4 26B A4B | $0.0037 | $0.06 | $0.33 | 262k | |
| 9 | Qwen3.5-Flash | Qwen | $0.0038 | $0.065 | $0.26 | 1M |
| 10 | Phi 4 | Microsoft | $0.0038 | $0.07 | $0.14 | 16k |
| 11 | GLM 4.7 Flash | Z.ai | $0.0038 | $0.06 | $0.4 | 203k |
| 12 | Mistral Small 3.2 24B | Mistral | $0.0042 | $0.075 | $0.2 | 128k |
| 13 | DeepSeek V4 Flash | DeepSeek | $0.0049 | $0.09 | $0.18 | 1.0M |
| 14 | Nemotron 3 Super | NVIDIA | $0.0049 | $0.08 | $0.45 | 1M |
| 15 | Ministral 3 3B 2512 | Mistral | $0.0052 | $0.1 | $0.1 | 131k |
| 16 | Qwen3.5-9B | Qwen | $0.0053 | $0.1 | $0.15 | 262k |
| 17 | Voxtral Small 24B 2507 | Mistral | $0.0056 | $0.1 | $0.3 | 32k |
| 18 | Llama 4 Scout | Meta | $0.0056 | $0.1 | $0.3 | 10M |
| 19 | Llama 3.3 70B Instruct | Meta | $0.0056 | $0.1 | $0.32 | 131k |
| 20 | Gemma 4 31B | $0.0067 | $0.12 | $0.35 | 262k | |
| 21 | MiniMax M2.5 | MiniMax | $0.0070 | $0.12 | $0.48 | 205k |
| 22 | Llama 3 8B Instruct | Meta | $0.0073 | $0.14 | $0.14 | 8k |
| 23 | Ministral 3 8B 2512 | Mistral | $0.0078 | $0.15 | $0.15 | 262k |
| 24 | GLM 4.5 Air | Z.ai | $0.0082 | $0.13 | $0.85 | 131k |
| 25 | Mistral Small 4 | Mistral | $0.0087 | $0.15 | $0.6 | 262k |
| 26 | Llama 4 Maverick | Meta | $0.0087 | $0.15 | $0.6 | 1.0M |
| 27 | Command R (08-2024) | Cohere | $0.0087 | $0.15 | $0.6 | 128k |
| 28 | Qwen3.6 35B A3B | Qwen | $0.0090 | $0.14 | $1.00 | 262k |
| 29 | Qwen3.5-35B-A3B | Qwen | $0.0090 | $0.14 | $1.00 | 262k |
| 30 | Llama Guard 4 12B | Meta | $0.0094 | $0.18 | $0.18 | 164k |
| 31 | Ministral 3 14B 2512 | Mistral | $0.010 | $0.2 | $0.2 | 262k |
| 32 | MiniMax M2.7 | MiniMax | $0.010 | $0.18 | $0.72 | 205k |
| 33 | Saba | Mistral | $0.011 | $0.2 | $0.6 | 33k |
| 34 | DeepSeek V3 | DeepSeek | $0.012 | $0.2 | $0.8 | 131k |
| 35 | Qwen3.6 Flash | Qwen | $0.012 | $0.188 | $1.13 | 1M |
| 36 | DeepSeek V3.1 | DeepSeek | $0.012 | $0.21 | $0.79 | 164k |
| 37 | DeepSeek V3.2 | DeepSeek | $0.012 | $0.229 | $0.343 | 131k |
| 38 | MiniMax-01 | MiniMax | $0.012 | $0.2 | $1.10 | 1.0M |
| 39 | GPT-5.4 Nano | OpenAI | $0.013 | $0.2 | $1.25 | 400k |
| 40 | Qwen3.5-27B | Qwen | $0.013 | $0.195 | $1.56 | 262k |
| 41 | DeepSeek V3 0324 | DeepSeek | $0.014 | $0.24 | $0.9 | 164k |
| 42 | DeepSeek V3.2 Exp | DeepSeek | $0.014 | $0.27 | $0.41 | 164k |
| 43 | MiniMax M2 | MiniMax | $0.015 | $0.255 | $1.02 | 205k |
| 44 | DeepSeek V3.1 Terminus | DeepSeek | $0.015 | $0.27 | $0.95 | 164k |
| 45 | Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image) | $0.015 | $0.25 | $1.50 | 66k | |
| 46 | Gemini 3.1 Flash Lite | $0.015 | $0.25 | $1.50 | 1.0M | |
| 47 | Gemini 3.1 Flash Lite Preview | $0.015 | $0.25 | $1.50 | 1.0M | |
| 48 | Qwen3.5 Plus 2026-02-15 | Qwen | $0.016 | $0.26 | $1.56 | 1M |
| 49 | GLM 4.6V | Z.ai | $0.017 | $0.3 | $0.9 | 131k |
| 50 | Codestral 2508 | Mistral | $0.017 | $0.3 | $0.9 | 256k |
| 51 | Qwen3.5-122B-A10B | Qwen | $0.017 | $0.26 | $2.08 | 262k |
| 52 | MiniMax M3 | MiniMax | $0.017 | $0.3 | $1.20 | 1.0M |
| 53 | MiniMax M2-her | MiniMax | $0.017 | $0.3 | $1.20 | 66k |
| 54 | MiniMax M2.1 | MiniMax | $0.017 | $0.3 | $1.20 | 205k |
| 55 | Llama 3.2 11B Vision Instruct | Meta | $0.018 | $0.345 | $0.345 | 131k |
| 56 | Qwen3.7 Plus | Qwen | $0.019 | $0.32 | $1.28 | 1M |
| 57 | Qwen3.5 Plus 2026-04-20 | Qwen | $0.019 | $0.3 | $1.80 | 1M |
| 58 | Mistral Small 3.1 24B | Mistral | $0.019 | $0.351 | $0.555 | 128k |
| 59 | Qwen3.6 27B | Qwen | $0.019 | $0.285 | $2.40 | 262k |
| 60 | Nova 2 Lite | Amazon | $0.020 | $0.3 | $2.50 | 1M |
| 61 | Nano Banana (Gemini 2.5 Flash Image) | $0.020 | $0.3 | $2.50 | 33k | |
| 62 | Qwen3.6 Plus | Qwen | $0.020 | $0.325 | $1.95 | 1M |
| 63 | Llama 3.3 Nemotron Super 49B V1.5 | NVIDIA | $0.021 | $0.4 | $0.4 | 131k |
| 64 | Llama 3.1 70B Instruct | Meta | $0.021 | $0.4 | $0.4 | 131k |
| 65 | Kimi K2.5 | Moonshot AI | $0.023 | $0.375 | $2.02 | 262k |
| 66 | DeepSeek V4 Pro | DeepSeek | $0.023 | $0.435 | $0.87 | 1.0M |
| 67 | GLM 4.7 | Z.ai | $0.024 | $0.4 | $1.75 | 203k |
| 68 | Devstral 2 2512 | Mistral | $0.024 | $0.4 | $2.00 | 262k |
| 69 | Mistral Medium 3.1 | Mistral | $0.024 | $0.4 | $2.00 | 131k |
| 70 | Mistral Medium 3 | Mistral | $0.024 | $0.4 | $2.00 | 131k |
| 71 | MiniMax M1 | MiniMax | $0.024 | $0.4 | $2.20 | 1M |
| 72 | GLM 4.6 | Z.ai | $0.025 | $0.43 | $1.74 | 203k |
| 73 | Mistral Large 3 2512 | Mistral | $0.028 | $0.5 | $1.50 | 262k |
| 74 | R1 0528 | DeepSeek | $0.029 | $0.5 | $2.15 | 164k |
| 75 | Nemotron 3 Ultra | NVIDIA | $0.029 | $0.5 | $2.20 | 1M |
| 76 | Nano Banana 2 (Gemini 3.1 Flash Image) | $0.031 | $0.5 | $3.00 | 131k | |
| 77 | Nano Banana 2 (Gemini 3.1 Flash Image Preview) | $0.031 | $0.5 | $3.00 | 131k | |
| 78 | Gemini 3 Flash Preview | $0.031 | $0.5 | $3.00 | 1.0M | |
| 79 | WizardLM-2 8x22B | Microsoft | $0.032 | $0.62 | $0.62 | 66k |
| 80 | Kimi K2 0711 | Moonshot AI | $0.033 | $0.57 | $2.30 | 131k |
| 81 | GLM 4.5V | Z.ai | $0.034 | $0.6 | $1.80 | 66k |
| 82 | GLM 5 | Z.ai | $0.034 | $0.6 | $1.92 | 203k |
| 83 | GLM 4.5 | Z.ai | $0.034 | $0.6 | $2.20 | 131k |
| 84 | GPT Audio Mini | OpenAI | $0.035 | $0.6 | $2.40 | 128k |
| 85 | Kimi K2 Thinking | Moonshot AI | $0.035 | $0.6 | $2.50 | 262k |
| 86 | Kimi K2 0905 | Moonshot AI | $0.035 | $0.6 | $2.50 | 262k |
| 87 | Kimi K2.6 | Moonshot AI | $0.040 | $0.66 | $3.41 | 262k |
| 88 | R1 | DeepSeek | $0.040 | $0.7 | $2.50 | 164k |
| 89 | R1 Distill Llama 70B | DeepSeek | $0.042 | $0.8 | $0.8 | 128k |
| 90 | Kimi K2.7 Code | Moonshot AI | $0.044 | $0.74 | $3.50 | 262k |
| 91 | Nova Pro 1.0 | Amazon | $0.046 | $0.8 | $3.20 | 300k |
| 92 | GPT-5.4 Mini | OpenAI | $0.047 | $0.75 | $4.50 | 400k |
| 93 | GLM 5.2 | Z.ai | $0.051 | $0.909 | $2.86 | 1.0M |
| 94 | Sonar | Perplexity | $0.052 | $1.00 | $1.00 | 127k |
| 95 | Grok Build 0.1 | xAI | $0.054 | $1.00 | $2.00 | 256k |
| 96 | GLM 5.1 | Z.ai | $0.054 | $0.966 | $3.04 | 203k |
| 97 | Claude Haiku 4.5 | Anthropic | $0.060 | $1.00 | $5.00 | 200k |
| 98 | Qwen3.6 Max Preview | Qwen | $0.064 | $1.04 | $6.24 | 262k |
| 99 | Grok 4.3 | xAI | $0.068 | $1.25 | $2.50 | 1M |
| 100 | Grok 4.20 Multi-Agent | xAI | $0.068 | $1.25 | $2.50 | 2M |
| 101 | Grok 4.20 | xAI | $0.068 | $1.25 | $2.50 | 2M |
| 102 | GLM 5V Turbo | Z.ai | $0.068 | $1.20 | $4.00 | 203k |
| 103 | GLM 5 Turbo | Z.ai | $0.068 | $1.20 | $4.00 | 262k |
| 104 | Qwen3.7 Max | Qwen | $0.070 | $1.25 | $3.75 | 1M |
| 105 | Mistral Medium 3.5 | Mistral | $0.090 | $1.50 | $7.50 | 262k |
| 106 | Gemini 3.5 Flash | $0.093 | $1.50 | $9.00 | 1.0M | |
| 107 | GPT-5.3 Chat | OpenAI | $0.116 | $1.75 | $14.00 | 128k |
| 108 | GPT-5.3-Codex | OpenAI | $0.116 | $1.75 | $14.00 | 400k |
| 109 | GPT-5.2-Codex | OpenAI | $0.116 | $1.75 | $14.00 | 400k |
| 110 | GPT-5.2 Chat | OpenAI | $0.116 | $1.75 | $14.00 | 128k |
| 111 | Sonar Reasoning Pro | Perplexity | $0.116 | $2.00 | $8.00 | 128k |
| 112 | Sonar Deep Research | Perplexity | $0.116 | $2.00 | $8.00 | 128k |
| 113 | Claude Sonnet 5 | Anthropic | $0.120 | $2.00 | $10.00 | 1M |
| 114 | Nano Banana Pro (Gemini 3 Pro Image) | $0.124 | $2.00 | $12.00 | 66k | |
| 115 | Gemini 3.1 Pro Preview Custom Tools | $0.124 | $2.00 | $12.00 | 1.0M | |
| 116 | Gemini 3.1 Pro Preview | $0.124 | $2.00 | $12.00 | 1.0M | |
| 117 | Nano Banana Pro (Gemini 3 Pro Image Preview) | $0.124 | $2.00 | $12.00 | 66k | |
| 118 | GPT Audio | OpenAI | $0.145 | $2.50 | $10.00 | 128k |
| 119 | Command A | Cohere | $0.145 | $2.50 | $10.00 | 256k |
| 120 | Command R+ (08-2024) | Cohere | $0.145 | $2.50 | $10.00 | 128k |
| 121 | Nova Premier 1.0 | Amazon | $0.150 | $2.50 | $12.50 | 1M |
| 122 | GPT-5.4 | OpenAI | $0.155 | $2.50 | $15.00 | 1.1M |
| 123 | Claude Sonnet 4.6 | Anthropic | $0.180 | $3.00 | $15.00 | 1M |
| 124 | Sonar Pro Search | Perplexity | $0.180 | $3.00 | $15.00 | 200k |
| 125 | Claude Sonnet 4.5 | Anthropic | $0.180 | $3.00 | $15.00 | 1M |
| 126 | Claude Sonnet 4 | Anthropic | $0.180 | $3.00 | $15.00 | 1M |
| 127 | Sonar Pro | Perplexity | $0.180 | $3.00 | $15.00 | 200k |
| 128 | Claude Opus 4.8 | Anthropic | $0.300 | $5.00 | $25.00 | 1M |
| 129 | Claude Opus 4.7 | Anthropic | $0.300 | $5.00 | $25.00 | 1M |
| 130 | Claude Opus 4.6 | Anthropic | $0.300 | $5.00 | $25.00 | 1M |
| 131 | Claude Opus 4.5 | Anthropic | $0.300 | $5.00 | $25.00 | 200k |
| 132 | GPT Chat Latest | OpenAI | $0.310 | $5.00 | $30.00 | 400k |
| 133 | GPT-5.5 | OpenAI | $0.310 | $5.00 | $30.00 | 1.1M |
| 134 | GPT-5.4 Image 2 | OpenAI | $0.430 | $8.00 | $15.00 | 272k |
| 135 | Claude Fable 5 | Anthropic | $0.600 | $10.00 | $50.00 | 1M |
| 136 | Claude Opus 4.8 (Fast) | Anthropic | $0.600 | $10.00 | $50.00 | 1M |
| 137 | Claude Opus 4.1 | Anthropic | $0.900 | $15.00 | $75.00 | 200k |
| 138 | Claude Opus 4 | Anthropic | $0.900 | $15.00 | $75.00 | 200k |
| 139 | Claude Opus 4.7 (Fast) | Anthropic | $1.80 | $30.00 | $150.00 | 1M |
| 140 | GPT-5.5 Pro | OpenAI | $1.86 | $30.00 | $180.00 | 1.1M |
| 141 | GPT-5.4 Pro | OpenAI | $1.86 | $30.00 | $180.00 | 1.1M |
Methodology
List prices in USD per 1M tokens from live market data (OpenRouter public catalog), refreshed every 6 hours. The per-request cost = input tokens × input price + output tokens × output price. Provider-side discounts (caching, batch) can reduce real costs — check each model page for cache pricing.